a philosophical reading of legal positivism
TRANSCRIPT
HAL Id: tel-01221770https://tel.archives-ouvertes.fr/tel-01221770
Submitted on 28 Oct 2015
HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.
A philosophical reading of legal positivismJuliele Maria Sievers
To cite this version:Juliele Maria Sievers. A philosophical reading of legal positivism. Philosophy. Université Charles deGaulle - Lille III, 2015. English. �NNT : 2015LIL30017�. �tel-01221770�
1
Université de Lille 3 – Sciences Humaines et Sociales
UMR 8163 – Savoirs, Textes, Langage
THESIS
Submitted for the Degree of
PhD in Philosophy at the Université de Lille 3
Advisor: Pr. Dr. Shahid Rahman
A Philosophical Reading of Legal Positivism
Juliele Maria SIEVERS
Université de Lille 3, July 16, 2015
Members of the Jury:
Pr. Dr. Patrice CANIVEZ Université de Lille 3
Pr. Dr. Sandrine CHASSAGNARD-PINET Université de Lille 2
Pr. Dr. Jan-Reinard SIECKMANN Friedrich-Alexander Universität (Erlangen-Nürnberg)
Pr. Dr. Matthias ARMGARDT Konstanz Universität
Pr. Dr. Shahid RAHMAN Université de Lille 3
2
3
4
5
To my brother, Joel.
(Para o mano.)
6
7
Table of Contents
1. Introduction……………………………………………………………………………………………………………………..1
2. The Author and his Theory……………………………………………………………………………………………….7
2.1 Three Schools of Thought, or: Why Legal Positivism? ………………………………………………….….8
2.1.1 Legal Realism……………………………………………………………………………………………………9
2.1.2 Natural Law Theory……………………………………………………………………………….……….10
2.1.3 Legal Positivism…………………………………………………………………………………………..…12
2.2 Why Kelsen? …………………………………………………………………………………………………….…….…….14
2.2.1 The Historical Context…………………………………………………………….……………………..14
2.2.2 The Pure Theory of Law (1960) ……………………………………………………………………..16
2.2.3 The General Theory of Norms (1979) …………………………………………….………………17
3. Preliminary Notions………………………………………………………………………………………………………..21
3.1 The Notion of Legal Validity………………………………………………………………………………..…………22
3.1.1 Defining “Legal Norm” …………………………………………………………………………………..22
3.1.2 Validity as Existence…………………………………………………………………………………….…24
3.2 Legal Science and its Object: Hume’s Naturalistic Fallacy………………………………………………27
3.2.1 Hume’s Naturalistic Fallacy…………………………………………………………………….……..28
3.2.2 The Relations between the Science and its Object…………………………………………30
3.3 Legal Conditions……………………………………………………………………………………………….…………..32
3.3.1 Imperatives and Norms………………………………………………………………….………………32
3.3.2 Causality and Imputation…………………………………………………………..…………………..34
3.4 Fictions………………………………………………………………………………………………………………………….35
3.4.1 The Hyerarchy of Norms…………………………………………………………………………….….36
3.4.2 The Basic Norm (Grundnorm) ………………………………………………………………………..38
8
3.5 Normative Conflicts and Logical Contradictions ……………………………………………………………41
3.5.1 The Problem of Applicability …………………………………………………………….…………..41
3.5.2 Conflicts between Norms ………………………………………………………………………………42
3.5.3 Derogation ……………………………………………………………………………………………….…..43
3.5.4 Temporality and Efficacy …………………………………………………………………………….…45
3.6 Practical Reasoning and the Rule of Inference ……………………………………………………….…….46
3.6.1 Dissipating the Confusion ……………………………………………………………………………..48
3.7 Final Remarks on this Topic ………………………………………………………………………………………….49
4. Legal Positivism: A Defense of the Scientific Method ……………………………………………………51
4.1 Positivism versus Natural Law Theories ………………………………………………………………………..52
4.1.1 Legal Validity and the Positive Law Perspective …………………………………….………56
4.1.2 Statics versus Dynamics ………………………………………………………………………..………57
4.1.3 Normative Content and the Natural Law Perspective …………………………….…….60
4.2 Understanding the Tension ……………………………………………………………………..……………………62
4.3 Overcoming the Tension ……………………………………………………………………………..……………….64
4.4 Final Remarks on this Topic ………………………………………………………………………………..………..65
5. Science and Method: The Naturalistic Fallacy ……………………………………………………………… 69
5.1 David Hume as a Model ………………………………………………………………………………………………..70
5.1.1 The “Ought” …………………………………………………………………………………….……………72
5.1.2 The “Is” …………………………………………………………………………………………………….…..74
5.2 One Example of the Fallacy: Jørgensen’s Dilemma ………………………………………….…………..75
5.3 Final Remarks on this Topic ……………………………………………………………………………..…………..79
6. The Non-Existence in Legal Science ………………………………………………………..……………………..82
6.1 Legal Science: Meaning and Particularities ………………………………………….…………….…………83
6.1.1 Some Examples …………………………………………………………………….……………………...84
6.1.2 Dichotomies in Kelsen’s Theory ……………………………………………………………………86
6.2 The Basic Norm as a Scientific Fiction …………………………………………….……….…………………..87
6.2.1 The Searching for Justification ……………………………………….……………………..……..88
6.2.2 Understanding the Fiction ……………………………………………………………..…………….89
6.2.3 A Little Help from Philosophy: Vaihinger’s “As-If” approach………………………….90
9
6.3 Where is the Basic Norm? ………………………………………………………………………………….…………91
6.4 Final Remarks on this Topic ………………………………………………………………………………..………..94
7. Normative Conflicts and Temporality in Law …………………………………………………………………96
7.1 Norms versus Sentences …..………………………………………………………………………………..………..97
7.2 Derogation ………………………………………………………………………………………………………………..….99
7.2.1 The Formulation of the Derogation Norm ……………………………………………………..99
7.3 The context of the Normative Conflict ………………………………………………………………………..101
7.3.1 Some Examples …………………………………………………………………………………………..101
7.4 Retroactivity ………………………………………………………………………………………………….……………103
7.5 The Basic Norm ……………………………………………………………………………………………………..……104
7.6 Final Remarks on this Topic ………………………………………………………………………………………..105
8. The Notion of Practical Reasoning (Part I) ……………………………………………………..……….…..107
8.1 Introduction …………………………………………………………………………………………….………….……..107
8.2 Understanding Jørgensen’s Dilemma ………………………………………………………..……………….109
8.3 First Attempts: Jørgensen and Ross ………………………………………………………….………………..109
8.3.1 Jørgensen’s Answer to the Dilemma ………………………………………….….……………110
8.3.2 Ross’ Answer to the Dilemma ……………………………………………….………….…………111
8.4 Kelsen’s Battle against the Dilemma …………………………………………………………………………..116
8.4.1 Correcting Jørgensen ……………………………………………………………………….…………117
8.4.2 Correcting Ross ………………………………………………………………………………..…………118
8.4.3 Kelsen’s Final Solution ……………………………………………………………..…………………119
8.5 Any Objections? Von Wright’s Deontic Logic ………………………………………………………………121
8.6 Final Remarks on this Topic ………………………………………………………………………………………..123
9. The Notion of Practical Reasoning (Part II) …………………………………………………..……………..125
9.1 Dialogs about Kelsen’s Solution ……………………………………………………………………….…………126
9.2 The Dialogical Approach to Logic ………………………………………………………………………………..127
9.2.1 Dialog, Validity, Truth and Justification ………………………………………………...…....128
9.3 The (Dia)logical Tools ………………………………………………………………………………………….………131
9.3.1 Preliminary Notions ……………………………………………………………………..……………..131
9.3.2 Set F Facts and the Problem of Justification …………………………….………………….134
10
9.3.3 The N Normative System ……………………………………………………….………………..….137
9.4 The Dialogical System DLLC2 …………………………………………………………………………………..…..141
9.5 Jérôme C.’s Guilt and Further Discussions ……………………………………..……………………..……142
9.5.1 Jérôme C.’s Example ………………………………………………….………………………………..142
9.5.2 Further Discussions …………………………………………………..……………………….….……145
9.6 Back to Jørgensen’s Dilemma and Final Remarks …………………….…………………….….……….147
10. Conclusion …………………………………………………………………………………….………………….……….149
Appendix ……………………………………………………………………………….………………………………….…….156
Bibliography ………………………………………………………………….…………………………………………….…..215
Abstract …………………………………………………………………………………………………………………………..220
Résumé ……………………………………………………………………………………………………………………………222
Acknowledgments …………………………………………………………………………………………………………..224
11
12
1
Chapter 1
Introduction
The present study represents a true effort in bringing closer two disciplines which have a long
history of mutual collaborations but also of remarkable tensions. By the way, the undeniable
connections between law and philosophy have already generated a hybrid discipline called
“philosophy of law”. This special branch aims to studying questions concerning the
fundaments of law, the justification of legal norms, the sources of law and also its fundamental
notions and concepts such as that of justice, for example, as well as the relations between the
legal and the moral field.
But the field of legal philosophy is a very vast one, since it tries to cover different issues
conforming the different aspects of the legal domain, such as constitutional law, international
law, contract law, and so on. Then, it seems reasonable that an interdisciplinary study must,
to begin with, find its specific place and have a specific objective and justification within this
legal context. Then, for the present study, we have chosen to place our inquiry in the field of
legal theory and, more specifically, the field of analytical jurisprudence.
Nowadays, the multidisciplinary feature seems to be a positive quality on the academic
context. But the interests in interdisciplinarity are not recent: Logic and Law are a perfect
example of two disciplines that were intimate related already when the period of Roman law
prevailed. Nevertheless, the results of the possible associations are not always positive ones,
and can sometimes be not only mediocre but also disastrous. Gabbay and Woods express the
idea of the necessity of a yet prudent perspective, as it follows:
2
“For what is interdisciplinary if not a kind of extraterritorial intrusion? It is
therefore hardly surprising that most interdisciplinary initiatives fail outright
or are dubiously graced by outcomes that are scorned, or merely ignored, by
the home disciplines.”
It seems that the motto to keep in mind when heading for such an undertaking is the idea
that, in the end, the “original” disciplines involved must also somehow profit from the results
that may be achieved. We hope it will be the case concerning the present study.
The scope of our investigation concerns the very basic questions involving the normative field,
and among them, the most philosophical one might be that of questioning what the exact
criterion would be to identify and circumscribe the notion of a legitimate legal norm, i.e. what
is that makes us recognize a determined norm coming from law as a binding norm on our
behavior. It is in the context of such a problematic that the concepts, notions and theses we
will analyze will be inserted.
This perspective allows us to develop the present study according to our general purpose of
offering a clear panorama of the possible relations between some legal and philosophical
concepts that are often used to deal with the aforementioned questions. This is also the
context by excellence of the so-called analytic jurisprudence.
In the context of analytical jurisprudence, the questions about the “nature” of legal norms will
be approached by three main schools of thought: the natural law tradition, legal realism and
legal positivism. Then, the various philosophers of law will have perspectives more or less
explicit in relation to the precepts defended by those three schools. Moreover, the discussion
goes so far in time that we can identify such ancient traditions such as Ciceron and the stoics
as defenders of the natural law theory perspective, together with the medieval and modern
versions defended by Thomas Aquinas, Samuel von Pufendorf and Gottfried Wilhelm Leibniz.
Then, more recently, Jeremy Bentham, Ronald Dworkin, John Austin, Herbert Hart and, of
course, Hans Kelsen can be seen as strict defenders of the legal positivist approach.
Within the frame of legal positivism, which is the legal context we decided to place our
investigations, there are numerous concepts and thesis concerning the legal normative field
that seem to be very fruitful for a philosophical discussion. On the same way, this
approximation with a philosophical perspective also seems to somehow promote confusion
and misunderstandings.
3
The aim of our work is to compare and analyze some of the notions that are “shared” by law
and philosophy, with the objective of promote a legitimate approximation of the two fields,
rather than artificial ones, like it is sometimes the case, namely when difficult and complex
questions such as the possibility of the application of logic to legal norms are simply taken for
granted.
Actually, the problem of the applicability will trespass all the chapters of the present study,
since our context concerning law will be provided by the legal theory of the Austrian jurist
Hans Kelsen (1881 – 1973), the main name of legal positivism, followed by Herbert Hart (1907
– 1992). It is well-know that Kelsen has a very strong interest in philosophy, attested by the
innumerable references in his works, mainly the last one, the General Theory of Norms (1979),
which is also the piece of work we will be focusing on in the present study. Besides, that
interest may be the reason why Kelsen is also known as the greatest philosopher among the
jurists, and the greatest jurist among the philosophers. But another important detail is the
fact that Kelsen’s interest is also marked by a methodological precaution: the almost obsessive
explanations about the difference between sentences and norms which, for not being neither
true nor false, cannot be corresponded or even compared to sentences, falling then
“automatically” outside the scope of logical application.
As we will explain further in the following chapters, legal positivism represents a suitable
frame for these conceptual analyses due to its methodological soundness, especially under
the Kelsenian perspective. By now, let us present the content of the present work as it is
divided and organized according to the following chapters:
In chapter 2, we introduce the author and the theory we will be using to frame our approach
to the legal domain. As we have already mentioned, we are basing our study in the positivist
legal theory proposed by Hans Kelsen, namely in his last book (published posthumously, six
years after his passing) called General Theory of Norms, of 1979. We will also present the three
aforementioned schools of thought in order to justify our preference for the positivist
approach, especially concerning the relations between law and philosophy.
In chapter 3, our aim is to provide a general overview of Kelsen’s project to Law. That was an
essential step to take, since Kelsen introduces several concepts and thesis that are original
and fundamental for the approaching of the problems we are dealing with. So, even if the
notions introduced in this chapter are going to be taken up again in the next chapters, it
seemed useful to present them as a general frame, since most of the concepts play an
4
important role also in the general question of the possibilities of relation between law (norms)
and logic.
In chapter 4, we insist on the relevance of legal positivism as an authentic background for the
comparison and analysis of the “shared” legal and logic/philosophic notions. This is
demonstrated by taking as a counter-example the case of the natural law tradition, which is
intrinsically related to subjective or metaphysical notions that fall outside the scientific,
objective profile of a legal scientific theory. At the same time, we propose an original approach
which could attenuate the methodological tensions between the two antagonist schools of
though.
In chapter 5, we take the case of a special kind of fallacy, introduced first by the famous
Scottish philosopher David Hume, and later named as the naturalistic fallacy. We investigate
the use that Kelsen makes of the notion of naturalistic fallacy in the context of is “pure” theory
of law, as a means to prevent the mixing of the legal science with the object it describes,
namely, the legal norms. In our view, Kelsen’s usage of the fallacy represents an effort towards
the approximation between the fields of law and philosophy.
In chapter 6, we analyze the presence of a notion that could be harmful to the thesis of the
“purity” of the positive legal theory, namely, the notion of Basic Norm as a fiction. We
investigate the correct place of that norm – legal system or legal science? – , an element that,
despite not having any prescriptive force or even any content, must still be considered “as-if”
it was a “regular” norm in order to perform its function. The philosophical treatment of the
notion of fiction as it is used by Kelsen is furnished by the philosophy of “as-if” of the German
philosopher Hans Vaihinger.
In chapters 7, 8 and 9, we study the two cases where the application of logic is especially taken
under consideration by Kelsen, namely, the problem of the application of the principle of non-
contradiction in the case of normative conflicts and the problem of the application of the rule
of inference in the case of the creation of a new norm.
In chapter 7, the case of normative conflicts is analyzed based on the function of the
derogation norm as a means to legitimately solve the conflict without appealing to logic. We
also consider the cases of the losing of normative validity (and, therefore, of existence) due
to temporal aspects, as well as the notion of efficacy, since it also plays a major role in the
cases when a norm is supposed to (or happens to) lose its validity.
5
In chapter 8 and 9, the special case of the application of the rule of inference to the situations
of normative creation is considered via the notion of practical reasoning. Accepting and
defending the legitimacy of practical reasoning and practical syllogism represents to fall in the
pit of puzzles and dilemmas such as the Jørgensen’s dilemma, which is also treated in those
two chapters.
Thus, in chapter 8, we start by explaining the dilemma, as well as two attempts to solve it.
Next, we present Kelsen’s critics to those treatments of the problem, as well as his own
attempt to an answer. His approach will be based on the important notion of modally
indifferent substrate as representing the neutral normative content to be inserted either in a
normative form or in an indicative one.
Chapter 9 represents a joint work with Dr. Sébastien Magnier, where we aim a treatment of
legal creation by means of a different kind of logical approach based of the dialogical approach
to logic. This treatment manages to respect and preserve the “purist” character of the
positivist legal theory as proposed by Kelsen. This is also our original contribution to the
context of problems such as Jørgensen’s dilemma.
Lastly, we present the conclusions that can be drawn from the set of conceptual analyses that
have been developed in all the previous chapters.
The Appendix will present the main notions concerning the logical tools used in chapter 9, as
well as three articles produced during the period of researches on the questions composing
the present study. Those articles, written in French, helped shaping the first ideas and
approaches to be later developed in the form of this doctoral thesis.
So, to sum up, as we perceive it, Kelsen’s approach represents a tireless effort in preserving
legal theory as an autonomous field regarding other sciences. Nevertheless, concepts like the
naturalistic fallacy and the fictions represent a mode of approximating the fields of philosophy
and law regarding the scientific means in which legal theory may use in order to approach its
objects, namely, the legal norms. However, at the same time, we observe Kelsen’s worries
about taking this relation for granted, especially concerning the particular logical notions of
the rule of inference and the principle of non-contradiction as purely logical ones (and never
juridical).
In front of such a context, the argument that can be sustained is that the philosophical notions
which are “welcomed” in the legal field are always supposed to perform in the scientific level,
6
so that they never specifically act over the norms in such a way as to modify the legal system
where they are directed to study.
7
Chapter 2
The Author and his Theory
“Hans Kelsen is unquestionably the leading jurist of the time."
(Roscoe Pound, 1934)
Perhaps one of the most well-known features of Hans Kelsen’s legal theory concerns his
attempts towards a “pure” theory of norms. Contrarily to other conceptions and theses1, the
notion of “purity” will be preserved along all of Kelsen’s works, more or less explicit
conforming the phase. The notion of “purity” as a methodological guideline also represents a
special feature of legal positivism when compared to other schools of thought, in the sense
that it presupposes rigid boundaries enclosing the legal normative level as, also, the object of
legal science.
In this sense, legal positivism establishes that only the “legal” can be source of Law. In the
same way, legal norms are going to be defined based only in the legal sources for their
creation, by no means appealing to elements outside that legal sphere. In this chapter we will
see how this positivistic approach contrasts to other schools of thought, namely the Legal
Realism and the Natural Law tradition.
The choice for this author as the name we have elected to represent legal Positivism is
explained by the very sound philosophical background traversing Kelsen’s writings: while the
Pure Theory of Law2 is marked by Kelsen’s Kantian phase, his later works advanced further
from this approach towards a more analytical perspective, were the preoccupations about the
1 Kelsen is also known for drastically changing is approach to certain points of his works, creating “phases” in his theory. This includes the problem of the application of Logic as well. 2 Kelsen, H. (1960).
8
influence of Logic in the legal practice have also grown more and more strong with the passing
of time, culminating in his posthumous work, the General Theory of Norms3, where the
problem of the application of Logic to norms is practically at the core.
2.1 Three Schools of Thought, or: Why Legal Positivism?
It’s a well-known fact that Kelsen’s most famous work is the Pure Theory of Law, whose second
edition of 1960 represents the canon of the legal school of thought called Positivism. Legal
Positivism represents, together with the Natural Law tradition and the legal Realism, the main
currents in legal theory.
The main objective of Hans Kelsen in the Pure Theory of Law is to establish a scientifically
oriented theory of Law, having the notion of “purity” as a leitmotif. Kelsen aims to “purify”
the legal method from every possible interference coming from external elements in the
description of the legal concepts. Those influences would possibly come from fields such as
politics, psychology or sociology, among others. For a positivist, it’s easy to notice the possible
influence of those disciplines in the other schools of thought, which would be, according to
Kelsen, a negative aspect attempting against an important feature of the very concept of
science: objectivity.
A good way to understand legal Positivism is by noticing how it differs from its “contestant”
schools of thought. Each of those schools will try to respond to questions such as what is the
right definition of a legal norm (as opposite to other directives such as orders or
commandments, or even to other kinds of also binding norms such as those coming from
moral systems). They also establish the foundations of a legal system, that is, they try to give
an answer to the question of when does an agglomerate of norms could be considered a full
normative system.
In the following sections we will present their respective answers to those fundamental
questions.
3 Kelsen, H. (1979).
9
2.1.1 Legal realism
Legal realism is the branch of study of legal norms mainly committed to the efficacy of the
norm as being the main element defining it. A criterion for defining a legal system must then
be based in its high effectiveness, meaning that the norms composing the system must be
systematically applied and observed in reality4.
Legal realism is until nowadays represented by the American tradition based on case law and,
as it might seem straightforward by the very definition, by the Common Law traditions. Even
if those traditions defend a neat autonomy from Morals, Religion (thus averting the Natural
Law conceptions) and Politics, the frontiers seem much more blurred when it comes to
Sociology, for example. The reason for that is because of the fact that, instead of explaining
and justifying how the subjects should behave, the theory seem to simply describe how they
actually behave: the legitimate norm is an efficient one, and the actual behavior serves as
model for the prescribed behavior contained in the legal norm. It is easy to notice that aspect
especially in the case of common Law, where the norm literally emerges from the custom or
from traditional and “standard” practices. A legal system is, by this perspective, simply
reduced to a collection of specific cases. If we may already introduce Kelsen’s terminology5,
the Realism doesn’t step out of the “Is” realm of the descriptions of the facts occurring in the
everyday life.6
The main name defending the precepts of legal Realism is that of the American jurist Oliver
Wendell Holmes Jr. (1841 – 1935), who established the fundaments of this school of thought,
being therefore also its founder7. Later on, further developments of this theory were
attempting to drawn more moderate approaches, but still fundamentally based on and
reduced to the American tradition8.
The critics that can be drawn to this approach are fundamentally based on the fact that legal
theory, differently from Sociology, must not explain facts in the same manner as natural
science does, i.e. by following the principle of causality that explains how nature “works”, but
4 The content of the norm must be observed by, for example, the citizens, and in the case when this does not happen, then the sanction linked to the lacking of the commanded behavior must be applied by the authorities. 5 To be further developed in the following sections of this chapter. 6 This level is separated from the “Ought” level of the prescriptions, the normative level of Law. 7 For a general and traditional overview of the theory of legal Realism, see Holmes, O. W. (1881). 8 The American Law Institute might be the most remarkable example among those initiatives.
10
it should rather explain how the legal norms are justified and legitimated. That means: the
object of a legal science is not the actual behavior of the agents, but the legal norms
themselves. This factor shifts the domain in which we are placing our discourse: a legal
science, a legal school of thought, must deal with the legal norms which are, in their nature,
prescriptions, imperatives, special kinds of commands. As a legal scientist, one has then to
step out of the “Is” level and step into the “Ought” level of normativity to approach their
subject of study, since norms act by qualifying those facts: they are not in the same level
anymore. The ultimate problem about Realism is that this normative aspect is ignored and
norms are dealt as social facts ruled by the principle of causality.
It is clear that Law is also a social phenomenon among others, and can evidently be studied
also by Sociology. But the results will not serve to explain the validity of a legal system or,
more precisely, its objective validity. Legal science must be able to “untie” the objective form
out of the multiple particular contents composing different legal systems, in order to
rationalize and construct a general objective theory. Departing from this general frame, the
jurist will then be able to approach and interpret the many instantiations belonging to
different jurisdictions.
2.1.2. Natural Law Theory
The Natural Law perspective has a long and solid tradition, drawn either by jurist as well as by
(famous) philosophers. The tradition goes back to the ius naturale belonging to the Roman
law systems. A good impression of what the term natural law meant in the context of Roman
law can be illustrated by the words of the Roman jurist Gaius (AD 130–180) when he says the
following:
“Every people (populus) that is governed by statutes and customs (leges et
mores) observes partly its own peculiar law and partly the common law of all
mankind. That law which a people established for itself is peculiar to it and is
called ius civile (civil law) as being the special law of that civitas (state), while
the law that natural reason establishes among all mankind is followed by all
peoples alike, and is called ius gentium (law of nations, or law of the world) as
11
being the law observed by all mankind. Thus the Roman people observe partly
its own peculiar law and partly the common law of all mankind.”9
It is interesting to notice that the ius gentium, in this context, was considered as a mere part
or an aspect of the broader ius naturale. Slightly different approaches of the notion of natural
law can also be found, among others, in the Stoic philosophical tradition10, namely in the
words of Cicero, when he says the famous words:
“True law is right reason in agreement with nature; it is of universal
application, unchanging and everlasting; it summons to duty by its commands,
and averts from wrongdoing by its prohibitions… It is a sin to try to alter this
law, nor is it allowable to repeal any part of it, and it is impossible to abolish
entirely. We cannot be freed from its obligations by senate or people, and we
need not look outside ourselves for an expounder or interpreter of it.”11
During the medieval period, the natural law tradition suffered from a strong influence of
Christianity, especially regarding the works of Thomas Aquinas (1225 – 1274). There is a
shifting of the source of Law, which no longer is established by Reason and marked by some
kind of “rational self-evidence” as in the Antiquity, but established by God and marked by the
“universal” values of good and just. Let us consider the following passage from Thomas
Aquinas:
“(…) there belongs to the natural law, first, certain most general precepts, that
are known to all; and secondly, certain secondary and more detailed precepts,
which are, as it were, conclusions following closely from first principles. As to
those general principles, the natural law, in the abstract, can nowise be
blotted out from men's hearts. But it is blotted out in the case of a particular
action, insofar as reason is hindered from applying the general principle to a
particular point of practice, on account of concupiscence or some other
passion, as stated above. But as to the other, i.e., the secondary precepts, the
natural law can be blotted out from the human heart, either by evil
9 Gaius cited by Winkel, L. (2004, p. 225). 10 And also during modern philosophy. This issue will be further developed in Chapter 4. 11 Cicero cited by Wood, N. (1988, p. 71).
12
persuasions, just as in speculative matters errors occur in respect of necessary
conclusions; or by vicious customs and corrupt habits, as among some men,
theft, and even unnatural vices, as the Apostle states (Rm. i), were not
esteemed sinful.”12
Later, Samuel Pufendorf (1632 – 1694), John Locke (1632 – 1704) and Gottfried Wilhelm
Leibniz (1646 – 1716) might all be cited as good examples of a naturalist legal perspective
based on theological foundations.
So, while legal realism is limited to the description of Law qua social phenomenon, natural law
engages with the evaluation and qualification of the reality. The criterion to legitimate and
define a norm is no longer based in its factual efficacy, but rather in a judgment of value, such
as the good or the just.
It is then easy to notice the methodological problems of such an approach in the context of
the construction of a legal theory: the natural law theory is based on a subjective criterion for
the definition of Law. It gets involved with ideologies and becomes doctrine rather than
science. It is important to remark that we are not defending the legal norms don’t have to be
just (neither is the legal positivism or the legal realism): the point is that justice cannot be the
measurement to define a norm, given that this definition has to be based in an objective
criterion.
2.1.3 Legal Positivism
Perhaps the very first remark to be made about the current of legal positivism should be that
it sees Law as a human construction, as something that didn’t exist before human’s will. This
first aspect will give then space to all further differences when comparing it with the previous
schools we have just presented.
The criteria for a norm to be defined as a legally binding one is, according to legal positivism,
its very own existence as a norm, that is, that fact of being “correctly” enacted or “posited”.
12 Aquinas, T. (2014[1265–1274] I-II qq. 90-106).
13
The specific existence of a norm it’s called legal validity. Only “posited” norms are valid, or are
consider as existent within the frame of a determined legal order. In order to be valid, a norm
must also be enacted by an authorized person, whose competence is attributed by the Law
itself. In the case of the Kelsenian theory, we might recognize the fact that this is indeed the
closest one can get to the notion of “purity”.
Natural lawyers will then attack the legal positivism by turning the attention to the
troublesome fact that, whatever content the norm will have, if it is still enacted and posited
according to the form predicted by the positivists, it will be a legitimate, valid, binding norm.
In this sense, the form “despises” the content. The function of the jurist will be, then, to
simply limit himself to impartially interpret this norm without questioning its value: if it’s good,
moral, just…
This problem continues to be evoked nowadays, and constitutes nothing more than a
miscomprehension of the finality of a legal theory. The problem of unjust or “bad” laws is a
legitimate one and should be dealt with most tenacity. But the point is that it is not the
function of the legal science to do so. The verification of the validity of a norm has nothing to
do with the subjective evaluation of its content and its practical consequences. That difference
is based on the classical fact-value distinction, constant in every serious scientific context.
Norberto Bobbio explains very well – as he always does – this problematic confusion when he
writes:
“In the same way, the fact that racial laws (to repeat the same example that
the modern moralists from law resort to) are iniquitous didn’t prevent that
they, unfortunately, existed, and that they have not only been valid, but also
efficient regarding the way they have been applied and have found
enthusiastic executors, along with an inert mass of accomplices. And if they
have existed, it’s clear that the problem of their validity, application or
interpretation was, in the orders they belonged to, a different problem than
that of their evaluation.”13
Nevertheless, one could still argue that the critics coming from the natural law theorists are
not about the laws already enacted in the past, but rather that the legal positivism should not
13 Bobbio, N. (2008, p.26). This passage is a personal translation from the version in Portuguese.
14
allow those laws to become valid or legitimate, so that the laws concerning slavery, for
example, should never have existed in the first place. In this case we should again remember
that the function of the jurist is not that of dictating or influencing the creation or the content
of norms, but rather to study and interpret those laws in an impartial manner. Admitting the
contrary is to give to the science an apologetic aspect that simply cannot be seriously
considered.
2.2 Why Kelsen?
It becomes now clearer why the legal theory of Hans Kelsen represents a good frame of study
of the relations between Law and Philosophy. More precisely, the philosophical aspects that
seemed to interest Kelsen are those belonging specially to the field of Logic. Kelsen’s aim is to
develop a theory for the knowledge of the legal material, to provide a criterion for giving to
the conjunction of multiple norms an objective rational unity.
The full commitment with the notion of “purity” regarding his approach to legal science will
later reveal itself to be a sober position about the common practice of application of logic to
the object of that science, namely, the legal norms.
2.2.1. The historical context
Hans Kelsen was born in Prague, in 1881. His family moved to Vienna in 1884, where Kelsen
studied Law at the University of Vienna. His doctorate thesis concerned the reading of Dante’s
Divine Comedy as a complex political metaphor. This approximation with Dante Alighieri’s
work, strongly marked by a catholic perspective, made the young Kelsen – originally coming
from a Jewish family – convert to Roman Catholicism in 190514.
In 1911, Kelsen accomplished his Habilitation in Public Law and Legal Philosophy, which is the
academic qualification allowing to hold lectures and to supervise doctoral candidates. Right
after this achievement, Kelsen was able to work as a full professor at the University of Vienna.
14 Seven years later Kelsen will marry and convert to Lutheranism.
15
It was during this period that he worked on the draft of the Austrian Constitution, enacted in
1920. Kelsen’s version is by the way the one still nowadays in effect.
During the 1930ies, Kelsen had grown more active in the political discussions of his time,
especially regarding the totalitarian profile of rising government. This was also the year when
Kelsen moved to Germany, working as a professor in the University of Cologne.
Three years later, the National Socialism rose in Germany and Kelsen immediately moved to
Geneva, in Switzerland. During this period, Kelsen also supervised the Habilitation thesis of
Hans Morgenthau (1904 – 1980). The two worked together in actively condemning and
criticizing the German National Socialism, due to what both became persona non grata in the
German territory during the entire prevalence of this regime. Morgenthau became a lifetime
friend of Kelsen, and later would become a famous name in the field of international politics,
working also as a consultant for the Kennedy administration in the United States from 1961
to 1963.
In the year of 1934, Kelsen publishes the first edition of the Pure Theory of Law. He continues
working in Vienna until 1940, when he decides to move to the United States to work at the
Harvard Law School. In 1945 he becomes full professor at the University of California,
Berkeley, where we would work for the rest of his life. There, he would later publish the
second edition of the Pure Theory of Law, in 1960, at the age of 78, achieving the work that
represents undoubtedly a major breakthrough in the field of legal theory.
In was during his mature years, especially after the 1960ies, that Kelsen’s interest in the
matters concerning the possibility of a logic of legal and moral norms started to grow. By the
year of his death, in 1973, Kelsen had collected a vast quantity of notes on this subject, which
were organized and published posthumously, in 1979, under the title of General Theory of
Norms.
Because of its special focus on the relations between logic and Law, and the very
“philosophically oriented” regard that Kelsen offers to his positivist approach to legal theory,
the General Theory of Norms is the main reference we will be using for the present study.
Nevertheless, the main notions and methodological precepts were, in their great majority,
already introduced in the second edition of the Pure Theory of Law. Let us see more carefully
how they were approached in those two main kelsenian works.
16
2.2.2. The Pure Theory of Law (1960)
The second edition of the Pure Theory of Law came to be the most known work of Kelsen, and
is where he introduces and explains many of the notions that will traverse the rest of his
writings. Nevertheless, some of the theses presented in the book were drastically modified
and corrected in the following works. This applies to the matter of the application of the logic
to norms (where Kelsen still admits, in this book, an “indirect” application by analogy, to be
later on, in the 1979’s publication, transformed into a full refutation) and also concerning the
notion of Basic Norm (that here is still seen as a hypothesis, to be later on regarded as a
fiction), for example.
But it’s in the Pure Theory of Law that Kelsen entirely commits himself to the methodological
profile that will mark his entire work. The notion of “purity”, the methodological abyss
between “Is” and “Ought”, the notion of legal validity as existence, the refutation of the notion
of practical reasoning, the hierarchical disposition of the norms in the legal system according
to a “pyramidal” figure, will all shape the further developments to be present in the later
General Theory of Norms15, the last work of his, where Kelsen philosophically profoundly
analyses his concepts and thesis while comparing them to traditional philosophic and legal
theories and vigorously responding to critics.
The Pure Theory of Law is also the piece of work that marks a Kantian phase of Kelsen’s
conceptions. Kelsen himself confirms this when he writes, in a letter of 1933 to Renato Treves,
that: “It is altogether correct that the philosophical foundation of the Pure Theory of Law is
the Kantian philosophy”16. With this perspective, Kelsen was actually trying to apply the
transcendental method to his theory. In the Pure Theory of Law, Kelsen tries to investigate at
a great length the following problem:
“Kant asks: ‘How, without appealing to metaphysics, can the facts perceived
by our senses be interpreted in the laws of nature, as these are formulated by
natural science?’. In the same way, the Pure Theory of Law asks: “How,
without appealing to meta-legal authorities like God or nature, can the
15 Kelsen (1979). 16 Kelsen (1998 [1933], p.171).
17
subjective sense of certain material facts be interpreted as a system of
objectively valid legal norms that are describable in legal propositions?”17
The answer given by Kelsen will be based on the notion of Basic Norm in its hypothetical
expression, a presupposition to be assumed by the legal scientist in his cognitive reference to
the norms of a determined legal system.
Thus, if in the Pure Theory of Law the philosophical questions are oriented towards epistemic
and methodological matters; in the subsequent General Theory of Norms it is the possibility
of logical relations between those legal norms that is the problem which seems to affect
Kelsen the most.
2.2.3 The General Theory of Norms (1979)
One could easily defend that the discussion around the topic of the possibility of a logic of
norms might be the main subject of the General Theory of Norms. If there are nevertheless
other important issues discussed in the book, as for example the different linguistic
expressions concerning the norms, their functions and their relation to science, all those
matters seem to lead to this main discussion, trying to give the arguments for the denial of
the possibility to apply traditional bivalent logic to legal norms.
Also, as the title suggests, Kelsen proposes a theory not only for legal norms, but also for those
of positive morals. Nevertheless, it does not seem that their identification is actually
straightforward. Even if Kelsen mostly illustrates his conceptions with examples coming from
the moral domain (especially from positive religious morality, such as Christianity), some legal
features are more difficult to be recognized in the moral field. One example might be the case
of normative conflicts which, in the moral domain, seem to be more similar to moral
dilemmas, which are treated completely differently from normative conflicts. The notion of
derogation sounds also very artificial when applied in the moral level. Moreover, another
remarkable difference concerns the notion of delict and sanctions: even if the morality also
“punishes” the behavior to be avoided, that conception of sanction is much looser, and less
17 Kelsen (1960, p. 202).
18
“institutionalized”, let us say, when compared to the legal one. Outside the religious level, the
notion of “authorization” also becomes problematic, which is clearly noticed when we
attempt to the fact that Kelsen always appeals to limited examples displaying paternal
relations, such as a father giving an order to his son, or of a teacher in relation to the student.
Back to the main question of the applicability of logic to norms, two paths are used in Kelsen’s
argumentation concerning this problem: the denial of application of the rule of inference in
the case of normative creation, and the denial of the application of the principle of non-
contradiction in the case of normative conflicts. The motto of the whole discussion lies on the
fact that norms, since neither true or false, are not able to be logically treated in the same
way as sentences are.
Kelsen offers this extreme negative position concerning the applicability as an answer to the
common practice of using law to deal with norms: “That they (the principle of non-
contradiction and the rules of inference) are applicable to norms of positive law has been
almost universally assumed in traditional legal theory”.18
Due to this critical perspective, this book offers us a perfect frame for our study. In our view,
the general hesitation on understanding and accepting Kelsen’s theory as well as his strong
position regarding the logical applicability is aggravated for two main reasons:
1) From the part of the juridical community: the discussions about the possible relations
between law and philosophy (mainly logic) are obscured because of the extent to which
Kelsen carried on his formalist project: Kelsen is interested in building a rigorous scientific
general theory, and his discourse might sound “empty” to the jurist who is interest in solving
the problems he/she faces in the everyday practice of law. Even if those problems have to do
with legal theory, it’s not in the work of Kelsen that the answers will be found. For example:
an important actual problem in Law can be found in the context of normative conflicts. The
jurist, or the lawyer, are interested in knowing, in the context of a situation of two different
norms being capable of rule one specific case, which of them should be picked up. The hidden
element behind the possible choice is the intention of the legislator or of the higher authority
while enacting the norm. It would be very useful to have then some kind of principle ruling
this kind of conflict or dilemma, in the same way that logic has its own principle of non-
contradiction. But Kelsen’s theory will nor answer to this kinds of “practical” questions: it will
rather give us a formal frame of the situation of conflict in law, while insisting on the fact that
18 Kelsen, H. (1979, p. 189).
19
a normative conflict should not be treated as a propositional contradiction, because two
conflicting norms have to be both valid in order to be in the situation of conflict in the first
place, which is very different from the case of contradictions between propositions, which
have the property of being true and false at the same time (A ¬ A). So, against the
jurist/lawyer expectations, not only the kelsenian theory will not give a practically applicable
answer, but will also insist on the fact that the possible answer given by logic is not applicable
nor legitimate in the legal field.
2) From the part of the philosophical community: the relations between law and philosophy
(mainly logic) are misinterpreted because of the fact that the legal concepts, once apparently
similar to the philosophical ones (validity, science, condition, fiction and many others) are
frequently confounded with their philosophical counterparts. That means that the concepts
receive their usual philosophical approach instead of being regarded within their legal
background (since legal validity is not the same as logical validity, science is not the same as
normative science, a legal condition is not the same as a logical condition or conditional and
so on).
In the frame of Kelsen’s General Theory of Norms, those difficulties are put into light. The
consequence is that, nowadays, this work of Kelsen is probably the least studied or
commented or referred to, despite its great philosophical richness. Kelsen developed, in this
piece of work, a thorough discussion between logic and law in order to set and fix the limits
on their legitimate approximation. Because of that, the relations are more limited, but also
more reliable and more sound.
For the jurist, the text and the problems that are treated seem to remain “too abstract”. For
the philosophers, only a few would agree that the “practical syllogism”, for example, where a
norm is derived from another norm together with its factual instantiation, fails as a model to
express the “creation” of new specific norms from general norms: they seem to take it for
granted, and see as almost natural the relation (and identification) of a valid norm and a true
proposition, probably because of the logical “equivalent” notion of logical validity, meaning
“true in all possible worlds”, or maybe because of the illusionistic evidence showed by
everyday examples such as those displaying the Jørgensen’s dilemma…
We propose, in the following chapters, a detailed study of the relations between those
apparently “shared” notions. The results might better explain why it is still so hard to put into
practice what is attested in the theory.
20
Our aim is not to establish a complete separation between the fields of law and philosophy
(and logic), but rather to establish a legitimate dialog, by respecting the autonomy of the two
domains and the impassable particularities that make them two areas of their own.
21
Chapter 3
Preliminary Notions
“A theory that explains everything, explains nothing”
Karl Popper
In order to develop the analysis and comparisons between Law and Philosophy in the manner
intended in the present work, we must first of all introduce the principal notions and
arguments proposed by the legal Positivism as it is seen by the legal theorist Hans Kelsen. The
criteria for the choosing of the concepts approached in this chapter are based on their
“similarity” or “approximation” to important concepts belonging to the philosophical field.
This represents somehow a problem, in the sense that, in the attempts of building connections
between the legal and philosophical field, those notions can be confounded and then
interpreted and applied in a wrong manner. Our attempt is precisely to clarify how far we can
go in those attempts, by elucidating and preserving the notions’ particularities. Besides, this
work might even – we hope so – reinforce the dialog between the two disciplines, since it
preserves a legitimate and thus reliable relation between them.
Moreover, the important aspect to be taken into account in this context is the problem of the
possibility of application of logic to norms. Most of the legal notions we will face have some
kind of “counterpart” in the logic tradition. This means that the notions share the same name,
but not the same meaning. The most remarkable is the case of the notion of validity. Legal
validity, in the tradition of legal positivism, has a drastically different meaning that the validity
22
studied by logic. This represents a major problem in the context of the matter of the logical
treatment of those norms, where this notion of validity is mostly misinterpreted.
However, before approaching those problems in the different chapters of the present study,
let us first introduce a general overview of the main concepts that will be involved in the
philosophical reading of legal positivism.
We will dedicate special sections in this chapter to, respectively, the notions of legal validity
as related to that of logical validity; the methodological separation between “Is” and “Ought”
and the naturalistic fallacy; the presence of fictions in law, specially the case of the Basic norm
in the legal positivism and the question of non-existence; legal conditions as opposed to logical
conditionals; the notion o temporality in Law in the context of conflicting norms and their
relation to contradictory propositions and, last but not least, Kelsen’s denial of the practical
syllogism (and the application of the rule of inference) as a means to express the normative
creation.
Those topics are therefore explained by taking as a background – more or less explicitly
depending on the subject in question – the problem of the possibility of application of logic to
norms.
3.1. The notion of legal validity
The first step when approaching the notion of legal validity in the positivist context is to
understand its identification with the specific existence of a norm. So, if validity is identified
with the mode of existence of a norm, let us see how the positivist tradition arrives at this
definition.
3.1.1 Defining “legal norm”
At first glance, it is easy to understand norms as being the meaning of something that happens
in reality and, after a specific interpretation, constitutes a norm. Let us consider the following
case, for example: if I want something to happen or someone to behave in a certain matter, I
23
express this will with an act, which Kelsen calls an act of will or an act of commanding. It is
simply the expression of one‘s “wanting” something to happen. It can be a linguistic act, a sign
drew on a board or a plank, a rising of a hand, a whistle… Then, this act has to be interpreted
in a specific manner – receive a legal meaning – in order to be understood as a norm. In this
sense, the linguistic act will be interpreted as an order from a Judge, for example; the sign in
the board will be interpreted as a limitation of speed in traffic, the rising of a hand will be
interpreted as a voting in a parliament, and the whistle will be interpreted as a permission to
cross the street. The norms which represent these meanings should not be identified with the
acts themselves, which are something factual occurring in reality and ruled by the principle of
causality. They merely “create” the norm, in the sense of positing it. It is easy to understand
how they are independent and should be separated from each other in this aspect when we
consider that the norm created by the act must continue to rule even when the act or the
willing of its author ceases to exist19.
Also, one important detail is that those acts will only be directly related to Law if they can
receive a legal interpretation. Otherwise they might be norms (in a weaker sense) but not
binding legal norms. The means to establish the possibility of this legal interpretation is given
by a more general norm, present in the legal order, which rules the situation where the “new”
norm emerges. Then, considering the abovementioned linguistic act of the Judge, for
example, the act is capable of receiving a legal interpretation because there is a “higher”
already existing legal norm that gives to the Judge the capacity, the authorization, to emit
such an act of will.
Departing from this scenario, we can now introduce the very basic positivist definition of a
legal norm as a specific meaning:
“The norm is the meaning of a willing or act of will and – if the norm is a
prescription or a command – it is the meaning of an act directed to the
behavior of another person, an act whose meaning is that another person (or
persons) is to behave in a certain way”.20
Moreover, in order to be classified as legal and binding, the meaning of the act of will in
question might not be simply subjective (which would then consist in a mere desire or whish
19 Or, more “dramatically” speaking: even after its author’s death. 20 Kelsen, H. (1979, p.2).
24
towards someone’s behavior), but it has to be objective. Objective, in this sense, means that
the addressees might recognize it as binding, and not only the person enacting the norm. The
element providing that objectiveness is the notion of empowering, or authorization. The
person enacting the act of will must be authorized (by Law) in order to produce binding norms
as the meaning of his act of will. At this point we can introduce this important precision made
by Kelsen:
“This is the only way in which the command of a highwayman can be
distinguished from that of a moral or legal authority. Generally: not every
Ought which is the meaning of an act of will is a binding norm. For instance: I
can will ‘Everyone is to marry upon reaching a certain age’. That is not a
binding norm for there is no norm of a positive moral or legal order which
empowers me to posit such a norm. The Ought in this case is only the
subjective meaning of my act of will, and not its objective meaning. It is only
when the Ought has an objective meaning, and so expresses a command, that
there exists a duty (i.e. a binding norm).”21
According to Kelsen, the statements are, by their turn, the meaning of acts of thought,
described in propositions which can be then evaluated as true of false. But this is no longer
the normative domain of the “Ought”, ruled by the principle of imputation, but merely the
indicative domain of the “Is”, ruled by causality.
The notion of objectiveness is also essentially linked to that of validity, since only when the
norm is the objective meaning of act of will it is also said to be a valid norm, i.e., to exist within
the frame of a determined legal order.
3.1.2 Validity as Existence
We have seen that the theory of legal positivism, contrarily to the natural law and the legal
realism traditions, aims to preserve the scientific objectiveness by using the notion of legal
21 Kelsen, H. (1979, p. 27).
25
validity as a criterion for defining a norm and a normative system. This notion is then identified
with the objective existence of a norm.
It is a curious fact that “validity” is a common term in Logic and in Law, because the notion
hides in reality two drastic different meanings. In Logic, we say from an argument that it is
valid due to the logical necessity of the conclusion, given the two premises. Also, a sentence
can be said to be valid, if it is true in all possible interpretations. In any case, what is interesting
is the fact that the notion of validity in Logic seems to be – in one way or another – somehow
linked to the notion of truth.
Now, let’s move back to the legal field. For a positivist theorist, the simple definition saying
that “When we say ‘A norm is valid’ we mean that a norm exists”, even if it seems almost
trivial, it is at the same time a fruitful ground for legal, logical and, mostly, philosophical
analysis. Let us see how it happens.
“Philosophically speaking”, we have to understand that, even if norms have a linguistic
formulation, even if they somehow give information, if they are written in the Constitution or
fixed in the walls of the public swimming pools of our cities, they are not to be treated as
normal sentences. What the positivist claim is saying is that norms have existence! In this
sense, when the legal norms exist, they are also, at the same time, “automatically” said to be
valid. Their existence is their validity.
This claim is fully committed with the idea of positing a norm which is, clearly enough,
essential to the legal positivism. Kelsen declares that “From the point of view of ethical and
legal positivism, the only norms considered to be objects of cognition are positive norms, that
is, norms posited by acts of will, and indeed, by human acts of will.”22 Kelsen distinguishes the
norms which are the meaning of those “real” acts of will from the norms which are the
meaning of a fictitious act of will – such as the Basic Norm. Fictitious acts of will won’t generate
norms with specific existence, but only with a fictional one. They do not really exist in the
frame of a determined legal order, and that is the sense in which the term existence must be
taking in this context.
Thus, validity is the ideal existence of a norm within the context of a determined legal order.
Therefore, considering for example the case of the Brazilian norm called the “Pelé” Law (Law
N. 9.615/98) enforcing that, by the year of 2001, (soccer) clubs can only sign a maximum five-
22 Kelsen, H. (1979, p.4).
26
year contract with a player, and allowing (soccer) clubs to create and promote their own
leagues; we must understand that this norm exists within the frame of the “Código Brasileiro
de Justiça Desportiva”, and it does not exist in the German Civil Code or the Brazilian Penal
Code.
There is still another factor that seems to increase the difficulty for the philosophical
understanding of legal norms: the validity is not even a property of a norm, in the same way
that we say from a proposition that it has the property of being true or false. Hans Kelsen
defines legal validity as the following: “Validity is the specific existence of a norm, an existence
different from that of a natural fact, and in particular from that of the fact by which it is
created. A norm decrees an Ought”23. And if the norm decrees an ought, it has a normative
function, that of commanding, empowering, permitting or derogating. To say that a norm is
valid is in fact a pleonasm, when norms are invalid, they are not binding, they are not part of
the legal set that constitutes Law, they are anywhere!24 Invalid norms are in fact simply
commands, orders, norms-to-be, without legal status.
This thesis also finds an echo in the philosophical tradition, namely, in the famous statement
of Kant when he says that existence is not a real (as opposed to a logical or syntactical)
predicate25. In this sense, if existence is not a real predicate or a property; existence is not a
characteristic which can be added to the concept of the subject. This argument was brought
up by Kant concerning the context of the ontological proofs for the existence of God. Kant
says:
“Both (the object and the concept) must contain exactly the same, and hence
when I think of this object as given absolutely (through the expression ‘it is’),
nothing is thereby added to the concept, which expresses merely its
possibility. Thus the actual contains nothing more than the merely possible. A
hundred actual dollars do not contain the least bit more than the hundred
possible ones. For since the latter signifies the concept and the former its
object and its positing in itself, then, in case the former contained more than
23 Kelsen, H. (1979, P.2). 24 There’s no authorized act of “positing” the norm anywhere. 25 Which, on its turn, finds an echo in Hume’s thesis concerning existence. After Kant, there is also the Frege-Russell view that existence is not a property of individuals.
27
the latter, my concept would not express the entire object and thus would not
be the suitable concept of it.”26
We can draw a bridge between the Kelsenian and Kantian perspectives in the sense that both
seem to be saying that the existence doesn’t “collaborate” in the explanation of a concept.
Taking the case of legal norms, it is completely redundant to say that a positive norm exists;
in the same way it is redundant to say that a positive norm is valid. The existence and the
validity are the very “nature” of the legal norm itself: it is only as existing and as valid that we
can grasp the notion of a positive legal norm.
Differently, “real” predicates such as good or just or efficient actually do add something else
to this existent norm, they are properties that tell us something about the norm. We are in
this case dealing with two different concepts which are put in relation to each other (for
example, that of justice and and that of norm), qualifying and thus broadening the meaning in
question.
3.2. Legal Science and its Object: Hume’s Naturalistic Fallacy
Now that we have seen what legal positivism follows as a methodological guideline, namely,
the notion of “purity”, we can analyze how that science approaches its object: the legal norms.
The case of legal science is a rather particular one, and very different from natural science.
Legal science describes norms, that is, it makes descriptions about prescriptions. Let us see
how that relation is articulated.
A norm can make part of a sentence which can be evaluated as true or false. Sentences about
norms, describing norms, are the content of a special kind of science, called normative science
or science of Law. Since the normative science contains true/false sentences about Law –
which is the overall set of norms, which can only be said to be valid – the limits between these
two spheres can be often erased, giving place to a number of confusions.
26 Kant, I. (1781/1998, 567 [B627=A599]).
28
3.2.1 Hume’s Naturalistic Fallacy
Facts of nature, causal links, scientific discoveries, they are all expressed by sentences, and
through our mental/rational evaluation they can be said to be true or false. All of these
elements belong to the realm of the “Is”. But positive norms are not given facts to be attested,
they need to be created, and they are created by intentional acts of human will. When a norm
is created as the objective meaning of an act of will, it is said to be valid, and decrees an
“Ought”. This specificity places the norms in another, particular realm, the normative realm
of the “Ought”, which is completely separated and not reducible to the indicative realm of the
“Is”.
This rigid separation is another constant methodological principle guiding the whole
positivistic theory. It represents a methodological “abyss” between the realms of the
normative – the “Ought” field – and that of the indicative – the “Is” field.
The mistake consisting in mixing those two separated domains represents an error that was
already pointed in History of philosophy, called the Naturalistic Fallacy. Even if the term was
introduced by G. E. Moore in his Principia Ethica of 1903, it is fully inspired on the traditional
problem of the is-ought separation approached by the Scottish philosopher David Hume (1711
– 1776) in his A Treatise of Human Nature of 1738. While the moorean version was more
focused on the problem of identifying good with the object qualified as good, the humean
version is more concerned with the confusion between the domain of the “Is” and that of the
“Ought” and the vanishing of the borders separating them.
The context of Hume’s criticism is the attempts to “naturalize” morals. In the legal context,
those attempts also exists, given that the problem of the fallacy arises precisely together with
the efforts to, for example, apply Logic to norms, that is, to treat norms as if there were natural
facts described by propositions. In Morals, the fallacy would be to consider “moral properties”
as “natural properties”, and Ethics as if it were a part of natural sciences. But in Law, the fallacy
would mainly be recognized in the efforts to treat legal norms as true/false sentences, instead
of the fruit of human arbitrary will.
The necessity and importance concerning a clear borderline between the two fields – factual
and normative – is explained by Hume himself: “Morals excite passions, and produce or
29
prevent actions. Reason itself is utterly impotent in this particular. The rules of morality,
therefore, are not conclusions of our reason”27.
This also clearly represents a refutation of the notion of practical reasoning, that is, of the
possibility of “creating” norms from reason or rational deduction. It is very important to
systematically reinforce the idea that, indeed, norms can also be the meaning of acts of
thought, which will give them their content, the information they’ll be giving to their
addressees (what to do or not to do, and the consequences following…). But that’s not yet the
moment when the norm is actually posited: this content must be seen as the “material” for
the subsequent act of intentionally willing it towards another one’s behavior.
Thus, the act of thought is simply a “first step” in the process of normative creation, an can
even be ignored, when we take in consideration the case, for example, of a parliament
member who votes for the approval of a norm without even knowing its content, i.e., what
the norm prescribes. The point is that the meaning of that action of the parliament member
was the creation of a norm, and it can have nothing to do with the enaction of a description
about someone else’s behavior, or an evaluation about it or any other mental or rational
outcomes.
The naturalistic fallacy is such a common argumentative mistake, that it is easy to recognize it
even in everyday examples, as well as in many other levels. Let us then see how the problem
of the naturalistic fallacy is identified by another authors, this time with a very contemporary
example:
“So, for example, Peter Singer takes E. O. Wilson to task for committing the
mistake of moving ‘from is to ought’ – specifically, for moving from the
premise ‘Our genes came from a common pool and will return to a common
pool’ to the conclusion that ‘we ought not to do anything which imperils the
human gene pool’ (Singer, 1981, p. 80). Singer correctly points out that this
argument is invalid as it stands – Wilson has neither made explicit, nor
defended as a moral value, the premise: ‘We ought not to do anything which
imperils the long-term survival of our genes’. Then, invoking nothing more
than Hume’s ‘unbridgeable gulf between facts and values’, Singer asserts that
‘ethical premises are not the kind of thing discovered by scientific
27 Hume, D. (1739/1985, p. 509).
30
investigation’ and hence ‘[n]o science is ever going to discover ethical
premises inherent in our biological nature’.”28
Once the necessity of clear and rigid divisions between “Is” and “Ought” is understood, we
can now also understand how the legal science approaches its object of study: the positive
legal norms.
3.2.2 The relations between the Science and its Object
Concerning precisely the relations between the science and its scientific object – here, legal
science and Law itself, it is important to insist on the fact that, even if the legal science deals
with valid norms, it doesn’t have any normative legal power.
The function of the legal scientist, or the jurist, is that of impartially describe, explain and
justify the concepts, theses and procedures followed regarding a determined legal normative
system. Within the context of a science, the jurists, while mentioning and repeating the norms
in their sentences about their validity, are ascertaining a fact to be verified by simply observing
the legal documents. This can only be the case in this context of legal positivism, where the
source of law is directly linked to the “belonging” of the norm to a determined normative
system.
Now, since the validity of the norms is also directly linked to the authority of the subject
enacting it, positing it, we can clearly remark that the jurist, while producing the theory, has
no authority to impregnate the scientific results with prescriptive power, with legal validity or
bindingness. The prescriptions of law will, in the context of science, have their validity
expressed by sentences, which will be true of false, depending on their correctness, which can
be attested by empiric verification.
One fine interpretation is the one that Norberto Bobbio makes of this unusual fact that the
normative science makes descriptions about prescriptions. Bobbio didactically explain how
28 Curry, O. (2006, p. 238).
31
the jurist is supposed to “perform” in his task of neutrally describing the norms of a given legal
order as the following:
“One can identify an oscillation and a superposition of two meanings, related
but not clearly separated:
a) the norms, a determined system of norms, are the point of view from which
the jurist, differently from the sociologist, considers the social behaviors; they
appear through a screen with a certain narrative structure, and the behaviors
are of interest once regulated and for the way they are regulated.
b) normative propositions are the result achieved by the jurist with his work
of verification, interpretation and systematization of a determined positive
legal order. Maybe we could say the same as the following: being the
normative science descriptive, it describes facts qualified as norms, i.e., by
norms, or it rather formulates ought-propositions by means of assertions. It
does not mean that those two meanings for “normative” are not conjugated
in the saying expressed many times by Kelsen, that the legal science is
normative while it describes what it is, but that in order to describe what
ought to be, it makes use of a determined legal system.”29
The “political” aspect behind that discussion is clearly the matter whether the legal science is
normative, i.e., whether it has the function of proposing “better versions” of the norms, or
still, whether it should correct norms or somehow “ameliorate” their formulation. Regarding
that question, Bobbio introduces the differences between “normative” and “prescriptive” to
explain that the normative science, despite the fact of dealing with norms – where the
“normative” term comes from –, does not make use of a prescriptive discourse, that is, their
descriptions don’t have the aim of changing the behavior of others, because they are
nevertheless still situated in the domain of the “Is”.
Of course that, in the everyday practice, things may not be so black in white. The myth of the
“impartial” researcher should, especially in the legal filed, not be entirely believed nor taken
for granted. The jurist has a critical reading when describing and interpreting the Law, and
perhaps this is even a positive aspect of his/her work. But what we should always keep in mind
is the purpose of Kelsen while interested in this kind of science, in providing a general formal
29 Bobbio, N. (2008, P. 59). This passage is a personal translation from the version in Portuguese.
32
theory: he is setting the way the jurist must precede, not the way he/she eventually ends up
proceeding.
3.3 Legal Conditions
Now that we have already explained how the definition of the legal norm is identified with a
meaning, namely the objective meaning of an act of will, it’s time to analyze how the norm is
linguistically formulated or constructed.
3.3.1 Imperatives and Norms
In order to introduce the discussion about the formulation of a legal norm, we can start by
approaching a strictly philosophical concept, namely, Immanuel Kant’s notion of “imperative
of skill”.
Since, to Kant, all imperatives express an ought, the imperative of skill will represents a “norm”
in the sense of a “rule” in order to reach an end via determined means. Kant says that: “All
sciences have some practical part which consists of problems of some end which is possible for
us and of imperatives as to how it can be reached. These can therefore generally be called
imperatives of skill.”30
So, according to Kant, an imperative of skill will be expressed in the following imperative form,
for instance:
“If you want water to boil, you ought to heat it to 100°C.”
But the point is that this means-end relation is clearly a causal relation. And causal relations
express a must, and not an ought, which is the normative component by excellence. The
means-end relation expressed by the imperative of skill is not a normative relation; it is not
expressed by a norm in the strict sense of the term, and does not posit an obligatory behavior,
30 Kant, I. (1785/2005, p.56)
33
it rather gives a recommendation, an advice concerning the reaching of a determined
objective.
These specifications concerning the Kantian notion of imperative are particularly useful to
notice that a norm is not as general as the notion of imperative or command. Indeed, the
definition of “norm” is somehow relied to the concept of a command or an order, but with
the important detail that this order must come from an authorized person, as an expression
– mostly in the form of an imperative – of a will.
Also, since this willing must come from a person authorized by the Law itself, there is always
a strict relation between the legal production and the legal power. A subjective act of will
consists only in a command, without legal force, it means the expression of a personal willing
towards a specific case, and that’s not what Law is about. Differently, an objective command
is not only the psychic event of the expression of a will. This can be seen in the case of a
testament, for instance. In a valid testament, the subjective act of will of the person making
the testament obtains its objectivity through the Law: once it is legally legitimated by an
authorized person or organ, the command of the person in question will remain valid beyond
his own existence, when he will no longer be able to express his subjective will. This
demonstrates the independency of the compulsoriness of the command from the subjective
act of will.
Kelsen illustrates this point by giving the example of a gangster’s demanding for money. When
the gangster asks you to give him all your money, this order, or the imperative he emits, has
not the same meaning as when the tax officer asks you for the money, namely, that the person
towards whom the order is formulated ought to render a determined amount of money. The
tax officer’s “request” is actually a biding valid legal norm, because it is based on the Law, and
the officer has the authority and the power given by the Law itself allowing him to perform in
the specific way he does. On the contrary, the gangster’s order represents only a subjective
act of will (even if it can be really appealing to call it binding, depending on the means he uses
to attain the ends), and it has the meaning of a command, but not of a binding or obligatory
valid norm.
Thus, while the command or the imperative are the expression of a desire, the norm is the
expression of a duty, of an “Ought” (Sollen). Let us now see in more detail how the relations
between the imposed behavior and the consequence linked to its disrespect are fixed in the
formulation of a legal norm.
34
3.3.2 Causality x Imputation
Another term belonging to the legal terminology and finding an echo in History of philosophy
appears when they we confronted to affirmations of the form: “All norms are valid merely
conditionally”, or also: “The general norms of positive morality and positive law are always
hypothetical”. In the present section we will analyze the notion of condition as it is expressed
in the formulation of norms.
Certainly, all positive norms link a behavior as obligatory or as forbidden under certain
circumstances. In the same way, when a norm prescribes a sanction, this sanction is seen to
be the consequence of the illegal act also expressed in the norm. This comes on the (very
simplified) form of “If someone commits an illegal act, then a sanction ought to be applied to
this person”.
In Philosophy, conditional sentences are logically evaluated by a logical connective called the
material conditional, which has the particularity of not specifying a causal relationship
between the antecedent and the consequent. In Logic, the “if … then” construction will only
be false when the antecedent is true at the same time as the consequent is false. So, the
sentence “If A, B" is false when A is true and B is false. All other combinations render the
sentence true, what yields a lot of bizarre consequences when we leave the formal level. For
example, any material conditional with a true consequent is true, and any material conditional
sentence with a false antecedent is also true (for example, “If I’m 10 feet tall, I play basketball
with my Martians friends on Saturdays”).
Given this context, when we say, together with Kelsen, that the legal norms are expressed on
the form of “If A is, then B ought to be”, then how is this supposed be comprehended under
a philosophical perspective?
Conditional legal norms cannot be treated as conditional sentences because the link between
the antecedent and the consequent is not linking two facts from the “Is” realm. What makes
the connection between antecedent and consequent in a legal norm belongs to the realm of
Law, the normative or the “Ought” domain, and the relation is then called a relation of
imputation.
35
The main difference between imputation and causality is that the imputation link needs to be
created, it has to be produced by a specific act of will for each and every norm, while the
causal link is completely independent from an agent. The causal link is what is expressed, for
example, on the rules studied by the natural sciences, such as the laws of physics, for instance.
In this sense, the laws of physics will continue to be true even if people don’t know about
them or would stop believing them: they are autonomous in this sense, they are “self-
governing”. On the contrary, legal norms are always the product of the human arbitrary will.
In this sense, the apparent “normativity” involved in the context of those natural laws doesn’t
really express that something ought to be the case, but they rather manifest some kind of
“regularity” or patterns that are systematized in and by the science.
So, in the imputation, there is no relation of necessity between antecedent and consequent,
because the link between the two states of affairs is always dependent on someone’s will or
decision. Consequently, in the norm “If someone commits a murder, he/she ought to be
submitted to criminal imprisonment”, no necessity link is present, since the existence of the
norm doesn’t avoid the fact that there are murders who are not put in prison. The “Ought”
link, once differentiated from the “is” link, that is, the duality between causality and
imputation, characterizes the methodological abyss which drastically separates those two
domains.
3.4 Fictions
For a person who is interested in the study of theories of law, but who does not have a formal
legal background, some particular kinds of normative formulations can seem to be very
confusing. Let us take the example given by Prof. Otto Pfersmann, when he talks about a
somehow “polemical”31 issue: legal fictions. Let us see the example, when Prof. Pfersmann
expresses the following norm:
31 The topic regarding legal fictions is an especially interesting one. But, unfortunately, a difficulty one can find in approaching this matter from a philosophical point of view (in the context of conferences or discussions with the Law professionals) is that there is a general hesitation from the part of the jurists to grasp how problematic and arbitrary the application of legal fictions can appear to be under a rigorous theorectical perspective. This is in part due to the fact that legal fictions are also highly “efficient” in the legal practice, which can give the wrong idea that their application is always theoretically sound. Many problems can be raised on the topic of legal fictions from a philosophical or logical perspective, and a whole thesis could be written by philosophers interested only in that matter.
36
“Only women have access to this swimming-pool, the man working as
swimming teacher is a woman in the sense of this statute.” 32
It seems strange to believe that this could actually be a valid norm. But this example shows
precisely the complete independence of the notion of validity in relation to what some would
say to be its correspondent: the notion of truth. In this example, the norm can be a valid one,
even if its content presents an impossible state of affairs, namely, a man to be a woman. This
norm is nevertheless perfectly understandable for their addressees (after an interpretation);
it can be, and actually it is respected by them in the same way that it can also be violated by
its addressees.
But these cases, where the fiction regards a – normally, contradictory – element present in
the linguistic formulation of the norm, i.e., in its very content, are not what interests Kelsen
and the theory he proposes. Kelsen is not interested in norms containing fictions in their
formulation, but rather in norms which are, themselves, of fictive nature. The presence of
fictions in the kelsenian theory is represented by the notion of Basic Norm, which is, in the
General Theory of Norms, considered to be an example of fiction by excellence.
But, before approaching the notion itself, it’s important to understand how norms are
“organized” in a positivist legal system.
3.4.1 The hierarchy of norms
Now that we explained that norms are valid when they are the objective meaning of an act of
will, let us see in detail how this normative creation happens.
Norms can be general of individual. Kelsen explains the difference between them in a very
clear manner:
The reluctance between the two fields regarding an open discussion on this specific topic could then, perhaps, be put aside. Due to this context, we will retain the discussion in this study to the notion of Basic Norm solely, and not in the legal fictions in general. 32 Pfersmann, O. (2004, p. 43). The passage is a personal translation from the original in French.
37
“A norm is individual if it decrees a once-only individually specified instance
of behavior to be obligatory. For example, the judicial decision that Schulze
the thief is to be imprisoned for one year. A norm is general if it decrees some
generally specified behavior to be obligatory. For instance, the norm that all
thieves are to be imprisoned.” 33
The content of such norms are always a determined human behavior34, i.e., the behavior
rendered obligatory by the norm. In this sense, it’s not entirely correct to say that a norm is
obligatory, because it is, instead, the behavior (actions or omissions) that is made obligatory
by the norm. By this perspective, individual norms are observed directly by the addressees,
while the general norms are observed only indirectly by them.
But the relation between the two kinds of norm is a direct “dependency” relation: the general
norm with corresponding content must be recognized as present in the legal order in question
(i.e., as “posited” or “written down” by the Legislator) so that the individual norm can be
created by the Judge for the specific case. Each individual norm needs its own act of will from
the Judge.
In this sense the individual norm must be grounded by a content-correspondent general norm.
Even if this process represents somehow an “individualization” of the general norm, no
deduction relation is established, since, as we’ve just mentioned, the Judge needs to enact a
brand new act of will whose meaning will be the validity of the individual norm. Without this
second act of meaning, now coming from the Judge, and which is completely separated from
the act of will of the Legislator concerning the general norm, no new legal norm is legitimately
created.
So, even if there exists a logical relation between the descriptions of the behaviors serving as
content in both general and individual norms (they are correspondent), there is no logical
relation between the validity of those two norms; i.e. between their existence. The individual
norm will only be valid if enacted by an authorized person, if it will be the objective meaning
of his/her act of will.
33 Kelsen, H. (1979, p.7). 34 We will approach the problem or the relation between form and content in the legal norm in the chapter dedicated to the study of the notion of practical syllogism (Chapters 9 and 10).
38
In this perspective, in order to give a general structure to the legal system, one can consider
the general norm as being in a “higher” position compared to that of the individual norm,
giving to this “lower” norm the reason for its validity (together with the act of will of the
Judge). Normally this structure is represented by a “pyramid” of norms, where the individual
norms serve as the basis.
The notion of Stufenbau is presented in the Pure Theory of Law and it’s taken for granted in
the following writings, especially in the General Theory of Norms, where it is barely even
mentioned. Nevertheless, this notion remains one of the most popular concepts of Kelsen’s
legal theory, i.e., it will mark Kelsen’s approach against Hart’s perspective of the normative
order according to different levels (primary and secondary legal rules – being the latest
represented by the rules of recognition, of change and of adjudication) as well as against other
more contemporary approaches considering the legal or moral orders as a complex normative
web.
In this sense, in order to answer to the question whether a given norm is a binding legal norm,
we have to consider two aspects: whether the content of the norm is grounded by a previous
– higher – norm and whether the person enacting the norm is authorized – also by a higher
competence norm, giving him/her the power to do so. The search for this grounding for the
normative validity should then meet an end when we arrive at the Constitution of a country,
for example. But then the further question still arises: “Which legal norm gave to the Legislator
of the first Constitution of a country the power, the authorization to enact it as a valid set of
norms?” There is no higher norm to ground it, no further element. Kelsen will suggest that it
can only be the presupposition of a higher norm that enables the normative chain constituting
the Stufenbau.
3.4.2 The Basic Norm (Grundnorm)
The notion of Basic Norm is one of the more intricate in legal theory. Since this norm is not
actually posited, it cannot be said to be valid in the same sense as all the other “ordinary”
norms composing the system, or to exist in the same way that they do, otherwise it would
also have to be present in the legal system, right above the Constitution, on the very top of
the pyramid, which would, again, perpetuate the problem of the lacking of a higher norm to
justify its own validity.
39
Kelsen argues that the Basic norm is a merely thought norm, i.e. it is the sense of a fictive act
of will. The Basic Norm should then be seen, as we will argue in the special chapter we
dedicated to the study of this notion35, as a methodological apparatus serving to give the
formal justification of the validity of the Constitution (and, subsequently, of the legal system
as a whole). Despite still having to be regarded as a norm (where the needing of the fictional
element comes from), this “norm” has a pure epistemological function: without it, the
collection of legal norms could not be seen as a coherent normative system passible of being
rationally described. It is only under the presupposition of the Basic Norm that legal science
can perform its function.
The notion of Basic Norm was invented by Kelsen and severely criticized by his
contemporaries. The legal positivist Herbert Hart himself has strongly attacked the notion as
being an unnecessary normative “duplication” and needless “complication” for the study of
legal positivism. For Hart, the bindingness of the constitution is simply rendered evident by
his “rule of recognition”. Differently from Kelsen’s perspective, Hart established a meta-level
for the rules of recognition permitting to recognize what is binding Law, to check if the steps
on normative creation were correctly taken (i.e. if the normative creation procedure was
correctly followed) to justify legal validity. However, Kelsen tries to avoid this upper separated
level of secondary rules in order to be able to keep positive law itself as the only source of
law (Stufenbau), preserving also the notion of “purity”. Hart’s normative perspective
grounded on the notion of rule of recognition clearly does not perform in the same way. Some
of the differences are that:
“(…) because the rule of recognition is a social rule, it is capable of being an
ultimate rule. It is ultimate in the sense that it does not exist in virtue of any
other rule. Its existence is secured simply because of its acceptance and
practice. The primary rules of the legal system, by contrast, are not ultimate
because they exist in virtue of the rule of recognition. The rule of recognition
validates, but is not itself validated.”36
Nevertheless, this approach contains its own difficulties concerning, for example, the possible
linguistic formulations of these rules of recognition, or even their relation with the naturalistic
35 Cf. Chapter 6. 36 Shapiro, S. J. (2009, p. 5).
40
fallacy, since it grounds legal validity in some kind of social commonly-known “standard”
practice, i.e., in the simple factual recognition of their authority.
The notion of Basic Norm might be more complex, but it seems fitter to meet the purposes on
establishing a general theory for legal positivism, since it preserves the theses and precepts
of the positivist tradition, which is not the case of Hart’s approach, at least considering the
hybrid conception of his rule of recognition.
Kelsen, very didactically, give us a very fine example, this time coming from the field of positive
morals instead of the legal field, that might help us explaining the place and function of the
Basic Norm in a positive normative system:
“For example: Paul comes home from school and says to his father: ‘My
classmate Hugo is my enemy, I hate him.’ Thereupon, Paul’s father addresses
an individual norm to him: ‘You are to love your enemy Hugo and not hate
him.’ Paul asks his father: ‘Why am I to love my enemy?’, that is, he asks why
the subjective meaning of he’s father’s act of will is also its objective meaning,
why it is a norm binding on him, or – and this is the same question – he wants
to know the reason for the validity of this norm. Whereupon his father says:
‘Because Jesus commanded “Love your enemies”.’ Paul then asks ‘Why is
anyone to obey the commands of Jesus?’; that is, he asks why the subjective
meaning of Jesus’s act of will is also its objective meaning, why it is a valid
norm, or – and this is the same question – what is the reason for the validity
of this general norm. The only possible answer to that is: Because as a
Christian one presupposes that one is to obey the commands of Jesus. This is
a statement about the validity of a norm which must be presupposed in the
thinking of a Christian in order to found the validity of the norms of Christian
morality. It is the Basic Norm of Christian morality, and it founds the validity
of all the norms of Christian morality. It is a ‘basic’ norm, because nothing
further can be asked about the reason for its validity. It is not a positive norm
(i.e. posited by a real act of will) but a norm presupposed in the thinking of
Christians, in other words, it is a fictitious norm.”37
37 Kelsen, H. (1979, p.254).
41
The tricky relations between norms and sentences about their validity on the context of the
presupposition of the Basic Norm need a more refined treatment, which will be delivered in
the chapter concerning the notion of Fiction in legal positivism38.
3.5 Normative Conflicts and Logical Contradictions
The interest of Kelsen in the problem of the applicability of logic to norms is marked by two
main questions that can be seen as instantiations of the general problem of the applicability,
namely:
1 – Could the principle of non-contradiction be applied to a situation of conflict between legal
or between moral norms?
2 – Could the procedure for the normative creation be expressed in a syllogism, and be ruled
by the rules of inference?
In the present section, we will analyze (1), leaving (2) for the next, following section of this
chapter.39
3.5.1 The problem of applicability
If we start the story by its ending, we can yet announce that Kelsen completely denies the
possibility of application of traditional bivalent logic to norms. The main argument will depart
from the fact that traditional logic can only be applied to true/false propositions – that is, it
presupposes truth values –, and the whole Kelsenian argumentation will be focused precisely
on expressing the differences between a valid norm and those true/false propositions. This
approach is also corroborated by Kelsen’s previous theses concerning the non-reducibility or
non-deduction between the realms of “Is” and “Ought”, as well as the differences between
38 Cf. Chapter 6. 39 Cf. Section 3.6.
42
acts of will and acts of thought and, of course, the preservation of the “purity” as a
methodological precept.
The position Kelsen takes with regard to the problem of the relations between Law and Logic
is clearly dependent upon his own definition of what a positive valid norm is. In this context,
Kelsen is trying to remove the focus from the possible deduction – in the case of normative
creation – or comparison – in the case of normative conflict – between the normative
contents. He then tries to put the focus on the problem of the relations between the validity
of the norms, that is, their legitimate existence. In this sense, we should face the problem with
regard to how those logical principles act on the validity of the norms in situation of conflict
or in the process of normative creation.
The point is that the validity is directly linked to the act of will positing the norm. This relation
cannot be grasped by logic, since it consist in a matter of the willing of an agent, an intentional
act arbitrary upon someone’s will towards someone else’s behavior. This last aspect is outside
the scope of the rules or principles of logic.
But let us go step by step and first consider how Kelsen deals specifically with the case of legal
normative conflicts, considering its possibility of being treated by the principle of non-
contradiction:
3.5.2 Conflicts between Norms
Concerning the definition of the normative conflict, it is important to note, from the very
beginning, that the situation of conflict is “located” in the legal modality concerning the
behavior contained in the norm. Professor Uta Bindreiter gives a clear explanation and an
example as it follows:
“(a norm conflict) is extant where the one norm renders a certain form of
behavior as obligatory, while the other norm renders obligatory a form of
behavior incompatible with the first.
Consider, for example, the following two norms, addressed to one and the
same norm subject:
Norm 1. Peter shall do A at 3 p.m. today.
43
Norm 2. Peter shall forbear from doing A at 3 p.m. today.
While norm 1 prescribes act A, norm 2 prescribes the forbearance therefrom,
that is, forbids the doing of A. Consequently, whatever Peter does (which
includes his being completely passive), he will inevitably violate one or the
other of these norms.”40
The main point to be stressed when defining normative conflicts is that both norms need to
be valid. In the same way, the sentences formulated by legal science describing the validity of
those norms will both be evaluated as true.
According to the terminology we’ve been using in our study so far, we can also recognize that,
while the norms are the meaning of acts of will, the sentences are the meaning or the
expression of acts of thought. The interest in stressing this aspect becomes clear when we
notice that, in the case of two contradictory sentences, one is evaluated as true and the other
as false. Also, the means to deal with normative conflicts is normally by derogation, when the
validity of one of the norms is abolished. But, in the case of the contradictory sentences, one
of them is false from the very beginning: the principle of non-contradiction does not “change”
or “abolish” the sentences’ truth values.
As the fact that contradictory sentences and conflicting norms have no similarities becomes
more and more clear, it becomes also necessary to deeply analyze how Law deals with the
situation of conflict, i.e., how the abolishment of one of the norms’ validity is performed, so
that the conflict is solved.
By preserving the notion of “purity”, Kelsen is interested in showing that no external element
is needed for the treatment of these “internal problems” of the legal domain, which may
eventually arise within the normative systems. But the main point is that law has its own way
to deal with the normative conflicts, let us see how:
3.5.3 Derogation
40 Bindreiter, U. (2002, p.135)
44
Derogation is not a principle in the same way as the logical principle of non-contradiction. It
cannot even be called a special legal principle. Derogation is one of the normative functions
among others (permitting, forbidding, empowering…). The derogation norm is created by the
legal authority only when in presence of a situation of conflict. This means that the derogation
norm is not one of the two norms in conflict, but a third norm. In this sense, it can be called a
“dependent” norm regarding the norms in conflict: the derogation norm has its own validity
as dependent on the validity of the norm it is supposed to repeal.
Thus, this derogation norm is not directed to the behavior of people with the intention to
regulate it, as the other norms of the legal order. Rather, it is directed towards the validity of
one of the norms in conflict, with the intention to abolish its existence. Therefore, the
derogation norm cannot be observed neither effective, nor even violated in the same sense
as the other “regular” norms. In order to introduce the concept of derogation, Kelsen uses
again an example of positive Morals, showing that is not only in the situation of normative
conflict that this norm can take place and be used by the authority of the normative system
in question:
“It can also occur in a positive morality: for example, in Christian morality,
when Jesus says: ‘Ye have heard that it hath been said, Thou shalt love thy
neighbor, and hate thine enemy. But I say unto you, Love your enemies…’
(Matt. 5:43-4). That means, I repeal the validity of the norm ‘Hate your
enemies’ and I posit the norm ‘Love your enemies’. The repeal of the validity
of the old norms and the positing of the new norm are two different acts,
which must be kept apart in the exposition of the morality in question.”41
Another aspect that is also important to notice is how misleading the legal principle of Roman
law saying that Lex posteriori derogate priori42 is, when it gives the idea that it is one of the
two conflicting norms (namely, the “newer” or more recent norm) that abolishes the validity
of the former, older norm. This legal principle is very common and indeed very useful in legal
practice, and it actually works as a presupposition in the mind of the Judge. With the
derogation norm it is similar: it is not permanently present in the legal order amongst the
41 Kelsen, H. (1979, p. 106). 42 The same is valid concerning also the principle “Lex specialis derogate lege generali”. A fine analysis of those principles in the context of conflicts is given by Kammerhofer, J. (2005).
45
other “regular” norms, but is taken for granted by the Judge when he/she is confronted with
a case of normative conflict. The derogation norm, as well as the principle Lex posteriori
derogate priori, is tacitly regarded as self-evident depending on the case in question, and
framed by the liberty of interpretation of the Judge. We consider its validity only as long as it
is directed to the validity of the norm-to-be-abolished, knowing that it will lose it at the same
time as the norm it abolishes.
3.5.4 Temporality and Efficacy
Another case in which the norm might lose its validity by derogation – besides the situation
of conflict – is when a norm stops being observed by its addressees or applied by Law, i.e.,
when it loses its efficacy with the time.
Efficacy is a term belonging to the indicative, factual level of what “Is”, because it is directly
linked to the fact of the observance of a norm or the application of this norm’s sanction
(punishment). Since general norms, according to Kelsen, are formulated by linking a sanction
to the disrespect of the behavior prescribed, the norm can be in this sense observed either by
the “citizens”, for example, as well as by the legal organs which have to apply the sanctions.
In the same way, it can lose it efficacy when it stops being either observed and/or applied
when it’s violated.
An example of a norm which has lost its efficacy by no longer being observed was the
Prohibition in the United States, concerning the sale, production, importation, and
transportation of alcoholic beverages, which remained valid from 1920 to 1933.43 The general
increasing of disrespect to the law with the passing of the years (allied to the development of
the black market concerning the sale of alcohol, together with the crash of 1929 and the
following Great Depression), lead to movements asking for its repeal. Finally, the law was
formally derogated in 1933.
On the other hand, the lacking of application of the sanction is also determinant. For instance,
an interesting fact is that, until a couple of years ago (2013), women were legally forbidden to
wear trousers in the city of Paris, in France. The law, which dates back to 1799, “required
43 Curiously enough, the consumption of alcohol was never prohibited during this period.
46
women to ask police for special permission to ‘dress as men’ in Paris, or risk being taken into
custody. In 1892 and 1909 the rule was amended to allow women to wear trousers, ‘if the
woman is holding a bicycle handlebar or the reins of a horse’."44 Of course, this law was not
being enforced during the last decades and, probably for that reason, or at least directly
related to that reason, women also stop observing it with the passing of time45. But, formally,
the law only stops existing in the legal documents after its derogation. That’s what happened
regarding this law in 2013, after the strong insistence of Mme. Najat Vallaud-Belkacem,
France's minister of women's rights at the time. But one has to recognize that, formally and
legally, until the year of 2013, that law was still a legitimate valid law within the territory of
the city of Paris.
With those examples we can see how, in the end, the validity of a norm – its existence – is
influenced by the notion of efficacy. We can say that a norm, in order to remain valid, must
be minimally – “by and large” – efficient. This represent a serious problem in the sense that it
“blurs” the limits between “Is” and “Ought” and takes the form of a naturalistic fallacy, when
the validity seems to be reduced to the notion of efficacy.
This major problem was recognized by Kelsen himself, who tries to somehow minimize it as
he defends that the efficacy would be a condition for legal validity, without being the reason
for it. He illustrates this thesis with a clever metaphor: “Just as the facts of being born and
being fed are conditions for human life without themselves being life, so the fact of being
posited and being effective are conditions for the validity of a norm without themselves being
validity.”46
3.6 Practical Reasoning and the Rule of Inference
44Seen in March 2015 at: http://www.telegraph.co.uk/news/worldnews/europe/france/9845545/Women-in-Paris-finally-allowed-to-wear-trousers.html 45 It is actually hard to fix a causal relation: the women stop observing the law because it was not being systematically applied anymore? Or is it the authorities who stop applying the sanction because the women stop observing the Law? The answer is vague because the lack of efficacy doesn’t happen from one day to another. It takes years, even decenies to observe the looseness of the relation between the laking of observance and the lacking of application. 46 Kelsen, H. (1979, p.140).
47
Besides the case of normative conflicts, another context where Kelsen considers the problem
of the applicability of logic to norms is that of normative creation, i.e., where an individual
legal norm is created by the Judge.
This question can be formulated in a broader sense, including not only legal norms, but also
moral norms and even “ordinary” imperatives. It consist then in the classical problem of
practical syllogisms, where norms are present in the premises and/or in the conclusion of a
syllogism.
People should love their enemies.
John is your enemy.
Therefore, you should love John.
Actually, as soon as we abandon the strictly legal field, we are able to remark how difficult it
would be to argue against the apparent validity of practical syllogisms as the abovementioned.
Nevertheless, the problem is that in the example above, we are “inadvertently” mixing, and
therefore taken for granted to correlation and even the identification of the truth value of a
sentence with the validity of a legal or moral norm47.
Besides all of the differences we already mentioned about sentences and norms, one of them
is of fundamental importance in the present context:
The statement, or sentence48 is the meaning of an act of thought. The norm is the meaning of
an act of will. Each act of will can “generate” one and only one norm. This means that the
norm in the conclusion of the syllogism cannot be derived from the norm in the major
premise, because it still needs its “own” act of will which is different and independent from
the factual act of will “generating” the norm in the major premise.
Thus, even if the content expressed by the norm in the conclusion is somehow correlated to
the content of the norm representing the first premise of the syllogism, it is still not possible
to establish any kind of logical relation between the validity, the specific legal existence of
those two norms. We need a brand new factual act of will regarding the norm of the
conclusion, and the fact of admitting the possibility of the simple logical deduction would be
47 We beg the reader not to confound legal validity with logical validity. 48 Kelsen seems to identify the notions of statement and sentence.
48
also ignoring the fact that there is no “authority” positing this new norm and legitimating its
existence.
Another way of explain this is by saying that there is no relation of necessity expressed by the
inference bar of the syllogism, as well as there is no preservation of the truth value or of the
legal validity, which themselves also have no correlation between them. A valid norm as a
major premise and a fact as minor premise won’t guarantee the validity of the norm in the
conclusion, because its validity is still dependent upon an act of will coming from an
authorized person.
Concerning this point we can also perceive a further difference between the legal norms and
the statements: while the validity of the legal or moral norm is directly dependent upon one
person’s act of will, the truth value of the statement is in nothing dependent upon one
person’s act of thought. This strong dependency relation between the norm and the act of will
is expressed by Kelsen with the formula “no imperative without an imperator, no norm
without a norm-positing authority, that is, no norm without an act of will of which it is the
meaning”.49
3.6.1 Dissipating the confusion
We can explain the apparent “evidence” of the logical validity of the argument by noticing
that it may actually come from the fact that the norm has a content which is indeed preserved
when passing through the inference bar of the practical syllogism. We know the content of
the norm which “is supposed to” be in the conclusion, and that is because of our acts of
thought, because of our rational reasoning about the content of the norms.
But still, that is different from enacting a legally binding norm, something which can only be
achieved by an act of will, coming from an authorized person. In the case of the “ordinary”
imperatives, there is no need of the binding aspect, but there’s still the need of a new act of
will directed to someone’s behavior. The point is that we are not allowed to step into the
normative level and assume that our reason is capable of positing norms, of establishing their
validity. The recognition of this fact represents not only a refusal and an attack to the notion
49 Kelsen, H. (1979, p. 234).
49
of practical reasoning, but also to any possible comparison or identification between the
outcomes of acts of will (imperatives and norms) and the outcomes of acts of thought
(sentences).
Thus, maybe the fact of using the right terminology might help on dissipating the confusion
around this misleading notion of practical syllogism. What actually happens when an
individual legal norm is created is that the existence, the validity of a general norm is
recognized (because its presence or not in a determined legal order is indeed a verifiable fact).
Then, when in presence of the case to be considered (previously expressed by the minor
premise in our example of practical syllogism), the Judge is capable of positing an individual
norm, via his/her act of will. Only then we have met all the necessary steps for legal normative
creation: the validity of a general norm with corresponding content and the act of will from
an authorized person.
Last but not least, another issue to be taken into account is that there must be always some
place left for the interpretation from the part of the Judge. There is no sense in structuring
this creative process in the scheme of a syllogism, since there is no question of preserving of
truth values nor legal validity, nor necessity when going from the “recognition” step to the
“positing” step.
3.7 Final Remarks on this Topic
The notions briefly presented in this chapter are the ones having a direct influence on the
analysis we aim to develop concerning the possible relations between legal positivism and
philosophy.
It is straightforward that there are several other legal notions that also deserve a thorough
philosophical analyses concerning their interdisciplinary pertinence. But, in our specific case,
where we take as background the Kelsenian problematic around the possibility of application
of logic to norms, together with the concepts and theses that this author makes use in order
to justify his approach, those are the most relevant notions to be analyzed.
50
So, after this general background of the terminology Kelsen established in order to explain his
scientific project concerning the positive legal systems of norms, we are now able to approach
each one of those notions separately.
51
Chapter 4
Legal Positivism: A defense of the Scientific Method
“The existence of law is one thing; its merit or demerit is another.”
John Austin50
In this chapter we intend to reinforce the idea of legal positivism as being the most suitable
context for a rigorous scientific approach to Law.
Since the general aim of our investigation is to develop a systematic study about the possible
relations between Philosophy and the study of norms, it would still seem reasonable to raise
objections of the like: “But Philosophy already deals with these problems since its ancient
developments!”. And that argument is actually fair enough. Besides, Aristotle’s moral
philosophy, for example, represents an effort to philosophically systematize the rules
concerning how people should rightly conduct their behavior in society. In a similar way, the
philosophical field of Ethics will analyze the norms of custom (ethos) according to concepts of
value such as the good, the evil, the just and so on, in order to provide systematized studies
about them…
Another interesting aspect of this philosophical context is that it actually seems to find the
real correlation of the legal field with the current of Natural law. The Natural law tradition
represents a normative system combining both legal and moral elements, i.e., it incorporates
50 Austin, J. (1832, p.157). John Austin is considered to be the founder of legal positivism.
52
moral values (good, just, valuable…) not only as a qualification but also as a criterion to define
what a legitimate legal norm is.
Then, the apparently genuine question arises: “Why not taking the Natural Law tradition as
the normative context to analyze all of those concepts shared by legal and philosophical
fields?”
We answer to this question by pointing to the fact that a legal theory based on Natural law
perspective will necessarily have to engage with concepts and thesis that fall outside the
scientific frame.
Nevertheless, our contribution to this discussion consists in the attempt to show that the
limits between those two schools of thought are less rigidly marked when we consider the
positivist theory as an empirical scientific approach for the description of a determined set of
positive legal or moral norms. This thesis will be explained in detail in the last section of the
present chapter.
4.1 Positivism versus Natural Law Theories
It is common-knowledge among the specialist in the works of Kelsen the fact that we can
identify several “phases” in the development of his legal theory51. The criteria for this kind of
periodization is based on several important aspects of Kelsen’s theory as, for example, the
influence of the Kantian philosophy over Kelsen’s general conceptions or the possible relations
between Logic and legal and moral norms52.
The first chapter of Kelsen’s General Theory of Norms53 is dedicated to the definition of norms
in the context of the legal positivism. While legal positivism is committed to defining a norm
by identifying it with its validity, a more intuitive attempt would maybe establish a
correspondence between the notion of norm and the content that is expressed by it, the
“something” which is commanded, or forbidden, or permitted. However, this kind of
51 The periodization of Kelsen’s legal theory is a very polemic issue. For the most important attempts and for an overview of the divergences, cf. Paulson, S. (1998) and Paulson, S. (1999), but also Heidemann, C. (1997) and Bulygin, E. (1990). 52 For a rigorous analysis of the specific problem of the applicability of Logic to the normative field according to three different phases, cf. Sievers, J. (2009). 53 Kelsen, H. (1979).
53
perspective is not proper to the legal positivism, but rather to Natural law approaches, which
consider the normative content as making reference to general moral principles that are
supposed to be followed.
For an exposition of the concepts developed in the context of legal positivism, a fine mode to
proceed may be to compare them with an “opposite” context such as the school of legal
naturalism. In this sense, we will try to display at what length the concepts belonging to those
two different approaches to law may still be obscure and hide problems on their very own
fields.
To start with it, as we have already seen, Kelsen develops his entire “pure” legal theory in the
frame of legal Positivism, which historically opposes itself to the Natural law tradition. Even
inside the tradition of positivism, Kelsen offers a new methodology, based on this concept of
“purity”, since he himself attacks his positivist predecessors by denouncing their tendency of
“old-fashioned” positivism, by identifying the norm with a fact, for example. For legal
positivists, every norm belonging to a legal system, and once enacted by a legal authority, is a
valid norm. This comes in contrast with the identification of the norm with an abstract moral
principle, as it was done by natural law scholars. Kelsen, in his turn, defends a rather unusual
definition of Law, namely, that “Law is Norm”. This claim is related to the well-known saying
making Kelsen’s enthusiastic devotion to the principle of purity much known at his times:
“Norm is Norm”54, Kelsen says. This statement, despite the fact of being a simple tautology, is
immediately understood by anyone implicated on the issues of the “defense” of legal
positivism: the definition of norm must be given without any reference to elements exterior
to the legal field itself.
But, in order to better understand the relation between Natural Law and Positivism when
trying to give a fair definition of what a norm is, let us go back in time and analyze the first
tensions regarding this definition in this classical passage concerning a dialogue between
Pericles and Alcibiades, described in Xenophon’s Memorabilia:
“It is said that Alcibiades, before he was twenty, had a conversation with
Pericles, his guardian and the head of the Athenian state, on the subject of
the Law.
54 Hart, H. (1998, p. 70). Concerning this saying of Kelsen, the whole passage of Hart making reference to it worth being quoted: “… towards the end of our debate, when upon Kelsen emphasizing in stentorian tones, so remarkable in an octogenarian (or in any one), that ‘Norm is Norm’ and not something else, I was so startled that I (literally) fell over backwards in my chair.”
54
‘Tell me, Pericles,’ Alcibiades asked, ‘can you explain to me what a law is?’
‘Of course I can,’ answered Pericles.
‘Then please do so,’ said Alcibiades. ‘For when I hear people praised for being
law-abiding, I am of the opinion that no man can rightly be praised in this way
if he does not know what a law is.’
‘What you want is nothing difficult, Alcibiades – to know what a law is. All
these are laws, all that the people in assembly approve and enact, setting out
what is or is not to be done.’
‘With the idea that good is to be done, or bad?’
‘Of course good is to be done, my boy, not bad!’ Pericles declared.
‘But if it is not the people, if, as in an oligarchy, it is a minority who assemble
and enact what is or is not to be done, what are these?’
‘Everything the ruling power in the state enacts with due deliberation,
enjoining what is to be done, is called a law,’ intoned Pericles.
‘Then if a despot, being the ruling power in the state, enacts what the citizens
are to do, is this, too, a law?’
‘Yes, everything a despot, as ruler, enacts,’ answered Pericles, ‘this, too, is
called a law.’
‘But what is force, the negation of law, Pericles? Is it not when the stronger
compels the weaker to do his will, not by persuasion but by force?’
‘Yes, that is my opinion,’ agreed Pericles.
‘Then everything a despot enacts and compels the citizens to do, without
persuasion, is the negation of law?’
‘Correct,’ said Pericles. ‘I retract my statement that everything a despot enacts
is a law, for without persuasion, his enactments are not law.’
‘And what the minority enact, not by persuading the majority but through
superior power, are we or are we not to call it force?’
‘I believe’, answered Pericles, ‘that without persuasion, whatever one
compels another to do – whether by enactment or otherwise – is force rather
than law.’
‘Then whatever the people as a whole enact, not by persuasion but by being
stronger than the owners of property, this, too, would be force rather than
law?’
55
‘Let me tell you, Alcibiades, when I was your age I, too, was good at this sort
of thing. We used to practice just the sort of clever quibbling I think you are
practicing right now.’
‘How I should like to have known you in those days, Pericles, when you were
in your prime!’”55
Thus, when Pericles ends up admitting that “law is simply an institutionalized expression of
the existing power relations”, he suddenly seems to recognize how strange and difficult it
would be to have to defend such a claim, and ends up admitting that it actually might not be
so.
But the only reason that leads Pericles to decline from the claim that law is what the despot
enacts is some kind of “intrinsic” belief supported by a moral background. The morality aspect
seems in this example to be clearly linked with the notion of “persuasion”: the citizens must
not to be forced; instead, there must be some sort of concordance and consensus over the
law. Therefore, while the initial intention of Pericles in the dialog is directed to a definition of
Law separating the legal field from morality (namely, by defending that the Law is the simple
expression of institutionalized power), Alcibiades seems, in his turn, to defend that this
definition must be explained in terms of morals in order to make sense or to be able to be
generally accepted.
Moreover, when regarding Natural law specifically, Aristotle’s Rethoric gives us a definition
based on the same kind of perspective. He makes reference to Sophocles’ Antigone, namely
in the passage where the sister buries her brother, something which, at those times, was
consider to be against the law. Nevertheless, the sister defends herself by saying that she
based her action in a higher law, a just law, something like “a law from above”. Aristotle
explains:
“We may begin by observing that they have been defined relatively to two
kinds of law, and also relatively to two classes of persons. By the two kinds of
law I mean particular law and universal law. Particular law is that which each
community lays down and applies to its own members: this is partly written
and partly unwritten. Universal law is the law of Nature. For there really is, as
55 Calhoun quoting Xenophon’s Memorabilia, I. ii. 40-6 (1944, p. 78-80).
56
every one to some extent divines, a natural justice and injustice that is binding
on all men, even on those who have no association or covenant with each
other. It is this that Sophocles' Antigone clearly means when she says that the
burial of Polyneices was a just act in spite of the prohibition: she means that
it was just by nature.”56
Those two examples show the intuitional difficulty in separating the legal filed from the moral
one. Then, in those cases, to give a definition of what a legitimate legal norm would be tends
to always point to the fact that its fair or goodness-directed content must be something to be
taken into account.
4.1.1 Legal Validity and the Positive Law Perspective
Kelsen’s theory of Law is a theory about the legal knowledge, the object being the norm itself.
In order to know and define what a legal norm is, we must isolate it from any external element
which we normally would associate with it, namely, justice, efficacy, sociological aspects, as
well as from the disciplines that could somehow be related, such as sociology, psychology or
ethics. Surely the field of Law involves all those aspects in its everyday practice, but the point
is that those elements must be excluded from the study and comprehension of law itself.
Concerning Ethics specifically, Kelsen insists that “the purity of the theory is to be secured
against the claims of the natural law theory, which… takes legal theory out of the realm of
positive legal norms and into the realm of ethico-political postulates”57. Thus, concerning the
definition of Law, Kelsen develops a particular approach to the investigation, turning the
question to: “How is positive law qua object of cognition, qua object of cognitive legal science,
possible?”, which is a very Kantian oriented perspective that marks one of Kelsen’s most
important and famous works, the Pure Theory of Law.
56 Aristotle. Rethoric. Book 1: Chapter 13. [1373b] 57 Kelsen, H. (1923, p. v).
57
Professor Stanley Paulson, one of the most important, if not the most important expert of
Kelsen’s theory in present times, interprets Kelsen’s aims regarding the nature of that theory
as the following:
“In putting his transcendental question Kelsen is not asking whether we
cognize legal material, whether we know certain legal propositions to be true.
Rather, he assumes that we have such knowledge, and is asking how we can
have it. To capture something of the peculiarly transcendental twist to
Kelsen’s question, we might ask: given that we know something to be true,
what presupposition is at work? More specifically, what presupposition is at
work without which the proposition that we know to be true could not be
true?”58
The answer clearly point to Kelsen’s intricate theory of the basic norm, which we shall not
discuss in length in this chapter59, but which will nevertheless help us to understand Kelsen’s
differentiation between Static and Dynamic principles ruling the normative systems of
positivism.
4.1.2 Statics versus Dynamics
Legal positivism identifies the norm with the normative validity, and not with its content.
Consider the following passage from Bruno Celano:
“According to Kelsen, validity is binding force. A norm has binding force if and
only if it ought to be complied with (obeyed). Saying that a norm ought to be
obeyed amounts to saying that one ought to behave as the norm prescribes.
So, for instance, asserting that the norm ‘Children ought to obey their parents’
is valid amounts to asserting that children ought to do what the norm
prescribes them to do. What the norm prescribes them to do is, however, to
obey their parents. Asserting that the norm ‘Children ought to obey their
58 Paulson, S. (1992, p.312). 59 Cf. chapter 6.
58
parents’ is valid, amounts, therefore, to asserting that children ought to obey
their parents. Generally speaking, whoever asserts that a norm ‘Op’, is valid is
at the same time asserting that Op, and vice versa (the statement “The norm
‘Op’ is valid” entails – and is entailed by – the statement ‘Op’). On Kelsen’s
conception of validity, therefore, asserting that a given norm is valid amounts
to (is logically equivalent to) asserting – i.e. iterating, accepting, endorsing,
assenting to, the norm itself.”60
This long passage also illustrates a very basic thesis of Kelsen, namely, the one saying that the
validity is not a property of the norm in the same way as the truth or falseness are a property
of the indicative statement. For Kelsen, the validity is identified with the norm, with its
existence as such. This important aspect gives rise to a remarkable consequence: to give a
definition of norm, to explain what is a legal norm equals to explain why the norm is valid, i.
e., to justify the validity of the norm. This procedure leads us to investigate the process by
which a legal norm is created.
Since positive legal systems are regulated by the principle of self-production and auto-
regulation, a norm must be created only when the act of will in question makes reference to
another previous valid norm. This process amounts to a regress which ends with a norm whose
validity is not based in a previous norm, which according to Kelsen can only be the Basic Norm.
Nevertheless, the support and foundation on a previous normative validity is still a necessary
but not sufficient step for the normative creation: there’s also a need for the unavoidable act
of will of an authoritative figure representing the Law, whose objective meaning will be the
norm. This additional fundamental step illustrates the importance of the notion of power or
authority in the context of legal positivism.
Thus, it is now clear that to understand what Law is, i.e., whether a prescription is a norm or
not, is to give a justification of its validity. Let us illustrate the searching for normative
justification by giving an example coming from the field of Literature: in the case of Sophie’s
Choice61, for instance, the command from the German officer to Sophie, forcing her to choose
which one of her two children would be sent to the gas chamber and which one would proceed
to the labor camp: is it a valid norm? What makes it binding (obligatory) to the addressee, to
60 Celano, B. (2000, p. 178). 61 Styron, W. (1979).
59
Sophie? In the context of legal positivism, despite its bizarre and abnormal substance, it is not
the normative content which defines the validity or the abidingness of the norm. According
to the precepts of positivism, the officer is indeed entitled to enact such a command; he is
authorized to enact this as a valid norm addressed to Sophie.
To Kelsen, at least in the Pure Theory of Law, normative systems can be classified as dynamic
or static. In a static normative system, the due behavior of the individuals is derived from the
normative content. The validity of such a norm is then based on the content of a more general
valid norm, and the regression goes until the most general principle serving as the content of
the most abstract norm funding a normative system. This kind of system is exemplified by
moral systems, and an example of the “most general principle” could be the Table of the Ten
Commandments in the case of Catholic religion, for instance. The Sacred Table is regarded as
“containing in itself” the content to be in all the other derived religious/moral norms, that is,
all the other possible norms have to have a content in accord to this general principle in order
to be legitimated, in order to be considered as “a part of” that specific moral system. To
Kelsen, at least in its early phase, Christian Religion and other positive moral systems
represent normative systems ruled by the static principle. This aspect also helps explaining
why, in religion or morality, we are normally our own “Judges”, that is, there is no need of an
authority to posit each one of the individual norms regulating and motivating one’s behavior.
We only consider the general posited principles and shape our own behavior according to
them.
But in a normative system ruled by a dynamic principle, the basic norm funding the legal order
provides no content to be derived, but only the attribution of legislative competence for the
creation of new norms. The basic norm here only provides the grounding for the validity of
the succeeding norms, and it is directed to the authority enacting the norms, by delegating a
power. In this dynamic process, no specific content is transmitted along the chain of
delegation, and the result of this purely formal process of rule producing is the astonishing
fact that any content can be a normative content, since the criteria ultimately lies in the
authoritative figure. To Kelsen, positive legal systems are the perfect example of this kind of
dynamic mode of operation:
“The norm system that presents itself as a legal order has essentially a
dynamic character. A legal norm is not valid because it has a certain content,
that is, because its content is logically deducible from a presupposed basic
norm, but because it is created in a certain way – ultimately in a way
60
determined by a presupposed basic norm. For this reason alone does the legal
norm belong to the legal order whose norms are created according to this a
basic norm. Therefore any kind of content might be law. There is no human
behavior which, as such, is excluded from being the content of a legal norm.
The validity of a legal norm may not be denied for being (in its content) in
conflict with that of another norm which does not belong to the legal order
whose basic norm is the reason for the validity of the norm in question. The
basic norm of a legal order is not a material norm which, because its content
is regarded as immediately self-evident, is presupposed as the highest norm
and from which norms for human behavior are logically deduced. The norms
of a legal order must be created by a specific process.”62
4.1.3 Normative Content and the Natural Law Perspective
Positive Law can fairly be defined as “a system which prescribes the rules of its own
development”63. Meanwhile, an even much generalized definition can be traced considering
natural Law:
“The first and most abstract notion that can be called ‘natural law’ is that
human beings are a certain kind of being, and the features of that being
should direct our understanding of how human beings should live. This
approach implies the existence of some sort of objective moral law knowable
through reason. It is implicit in what are perhaps the most basic intuitions
giving rise to natural law, namely, the sense that there must be some general
standard in light of which it is possible to judge human laws or conventions.
The classic instance of this idea is found in Sophocles’ Antigone, in which a
sister disobeys a law by burying her brother, and claims a warrant in higher
law for so doing.”64
62 Kelsen, H. (1960, p. 198). 63 Silving, H. (1955, p. 489). 64 Wolfe, C. (2003, p.38).
61
For the Thomist perspective of natural law, the evidence of a moral oriented principle guiding
the creation and justification of the norms is even stronger:
“…what to the Thomist is the basic, primary and self-evident principle of the
natural law (…) is variously phrased as: follow your nature; act according to
your proper end; maintain order; act for your rational end in conformity with
your total nature; bring your essential being to completion; follow your
rational inclinations; do good and avoid evil (meaning what is good and bad in
light of human nature), etc. In other words, act according to reason.”65
These passages only reinforce the fact of Legal Positivism and Natural law as having a long
history of conflict in legal Philosophy. The main argument of Natural law defenders points to
the necessity of a moral grounding for the norms: a norm, in order to be valid, must be a fair
norm, a quality which is achieved by the derivation from a general principle given by God or
Nature or, due to its evidence, by Rationality. This claim covers not only philosophical, but also
ideological allegations. It implies that Natural law involves a notion that would never and could
never be accepted in the positivist tradition: the subjective and therefore non-scientific notion
of belief66 as a criteria for legitimacy, represented by moral principles for acting, when not
religious reasons for acting.
Therefore, for a Natural lawyer, the only definition of authority will be linked to the normative
power emanating from a “superior” abstract and very general principle which will forcedly
impose a determined value. Kelsen characterizes Natural Law Theory as the following:
“If the ‘nature’ in which the norms of Natural Law are immanent is the nature
of man, and if we consider that man’s nature – unlike that of animals – consists
in human reason, then Natural Law appears as the Law of Reason. Since
reason is the faculty of thought and knowledge, the norms of the Law of
Reason present themselves as the meaning of acts of thought: they are not
willed norms but thought norms. Now there is such a thing as a merely
thought norm (as opposed to a positive norm posited by a real act of will); but
it is not the meaning of an act of thought but rather of an act of will which
65 Natural Law for Today’s Lawyer (1957, p. 479). 66 For a complete overview, cf. Reynolds, B. (1993).
62
does not occur in reality but is imagined or thought of in the same way as we
can imagine or think of anything which is possible but does not exists in
reality.”67
Therefore, the main point is not even that the natural law doesn’t fit the definitions traced by
the positivist tradition, as anyone would expect, and by so falling out of the scope of what law
is or is supposed to be according to legal positivism. The more basic point is rather that the
natural law doesn’t even step out of the descriptive field, where norms are simply thought
because, instead of being produced by human will, they are only attained or reached by
human reason, by an act of thinking.
What we actually observe here is a clear negation from the part of Kelsen of the notion of
practical reason. Clearly, the normative field, the domain of the regulation of human behavior,
cannot be ordered and ruled by particular statements derived from general abstract
ideological metaphysical/theological formulas such as “do good and avoid evil”, which are
strange to the scientific aims of legal cognition itself.
4.2 Understanding the Tension
Considering the Kelsenian distinction between static and dynamic principles ruling the
normative creation, the Natural law kind of normative systems were classified as ruled by a
static principle (as moral orders), while the positivist normative systems are clearly ruled by a
dynamic principle.
But then, since the Natural law norms seem to be inserted in a static normative system, where
we take as point of departure a general widely-accepted abstract principle and derive every
other norm from their content, there is no space left for the acts of will in the enactment of
the derived norms. Therefore, these moral norms are not wanted norms, but merely thought
norms, the result of a pure mental process of logical derivation. They are not the meaning of
67 Kelsen, H. (1979, p.6).
63
acts of human will, but are the result of rational deduction from the content of a general
principle.
The positive legal system, on the contrary, is based on a dynamic principle which allows the
agent to enact a valid legal norm, which is created as willed, more than merely thought. We
have nothing to say about the “quality” of the content of such norms, but only about the
legitimacy or not of their validity. It is the process of normative creation which is regulated by
the norm itself, internally to the system at stake, which preserves also the kelsenian “principle
of purity”.
Many find in this approach an opportunity to criticize legal positivism as a fruitful field for
totalitarian or anti-democratic regimes. However, Kelsen has never denied that a legal system
indeed must be revised and criticized when in need. But the point is that such an activity is
not the task of legal theory. It is only the legal science and its methodology which must be
“purified”, and it is the legal scientist who should isolate the legal norm from every “external”
element, such as justice for example, in his study, analysis, explanation and interpretation of
the legal norms. This methodological “separation” is very well illustrated in a passage coming
from Kelsen’s last, farewell lecture at the University of California in Berkeley on May 27th,
1952. The context is a by that time 71 years old Hans Kelsen, once forced himself to leave
Germany during WWII:
“And, indeed, I do not know, and I cannot say what justice is, the absolute
justice for which mankind is longing. I must acquiesce in a relative justice and
I can only say what justice is to me. Because science is my profession, and
hence the most important thing in my life, justice, to me, is that social order
under whose protection the search for truth can prosper. ‘My’ justice, then,
is the justice of freedom, the justice of peace, the justice of democracy – the
justice of tolerance.”68
The fact of the lacking of objectivity for the concept of justice doesn’t hinder the possibility
for it to be an ideal to be searched for. The point is simply that this searching cannot be
performed in a scientific frame. It is not possible, for example, to propose a strictly scientific
theory that would ground some of the more basic Natural law principles such as Lex injusta
68 Kelsen, H. (2008 [1957], p. 24).
64
non est lex, defended by St. Augustine and St. Thomas Aquinas, simply because the notion of
justice is subjective. Then, even if we all can eventually agree with some general idea of justice,
the real problems would arise later, when this general concept will be supposed to be applied
in the many different cases…
In the context of positivism, the closest one can get to the notion of justice would be, in this
sense, an almost formal notion of equity, meaning simply that similar cases must be treated
similarly. But we must recognize that this formulation represents a very abstract, nearly empty
or useless precept.
4.3 Overcoming the tension
Alf Ross gives a good oveview of the tensions between legal positivism and natural law when
he explains that:
“This conflict is often treated as the most fundamental issue in legal
philosophy, dividing the field into two hostile and irreconcilable camps.
Positivists characterize natural law doctrines as beliefs based on metaphysical
or religious ideas incompatible with the principles of scientific thought.
Proponents of natural law theory, for their part, accuse their antagonists of
failing to understand the realm of spirit and value, a realm that is real enough,
although it cannot be discovered or described by means of sensory
experience. Natural lawyers have even gone so far as to accuse the positivists
of moral torpidity, and of complicity in the abominations of the Hitler
regime.”69
As we understand, the critics coming from each of those fields and directed to each other are
actually not based in an exact recognition of the methodological functions established by their
corresponding legal theories. For that reason they seem to be somehow mixing the theory
with the object to be studied. Let us see how that happens in more detail.
69 Ross, A. (1961, p. 46).
65
It is not that evident that the manner in which the Natural law jurists regard legal positivism
is completely incompatible with their own view. Rather, it seems that, for a Natural law
scientist, the positivist approach is merely “incomplete”70, in the sense that it ignores the
necessity of grounding for the basic precepts of a normative system in those most
fundamental values that would also ground the life in society in a safe and constructive, rather
than dangerous and destructive, “universal” intentions. Moreover, History has repeatedly
taught us that, after a “high dose” of legal positivism, there were always the need to make
reference to Natural law as the only means to re-establish the social order. Both systems
seemed to have “interacted” in this sense. It is easy to recognize that aspect once we take for
example the cases like the Nürnberg trials after World War II, where the war crimes
committed were then qualified as crimes against humanity or as violations of the human
rights.
At the same time, we believe that the positivist should not be blamed for the abnormal results
that may come out of a positivist legal system, since also the natural lawyer is not safe from
the possibility of the development of some abnormal possible beliefs that may be defended
in the name of the Natural law system. Religious fanatical is one of the most frightening
aspects of society. Nowadays, for example, the Brazilian federal congress is taken by
numerous right-wing congressmen coming from military and/or religious background and
defending clearly homophobic, racist statements and even expressly inciting hatred and
violence against women and homosexuals. All of that is defended in the name of the interests
and the preservation of the idea of the ideal “traditional Brazilian family”. Those politicians
are also the most voted ones during the Brazilian elections. So, this notion of idealized
“common good” underlying such offensive and harmful claims, expressly mixing religious and
personal beliefs in the interest of establishing laws corresponding to it, is it really humanistic?
At the end of the day, what would be the ultimate criterion for defining the “natural” in
Natural law?
4.4 Final Remarks on this Topic
70 But the relation is not transitive: for the positivist, the natural lawyer’s approach is simply wrong.
66
The positivistic perspective will deny the legitimacy of natural law in the same way it denies
morality as grounding for legal validity, because moral principles cannot be formally known,
in the sense of being scientifically objectively defined.
But one thing that can be firmly stated is that, concerning the method for the description of
the legal system, the frontiers between the two approaches are notably less marked. For the
positivist scientist, or jurist, the task to be performed is the description of the fact of the
existence of the norm, because the presence of a norm in a legal document is indeed a fact
that can be attested, in the same way as the enaction of an individual norm in the frame of a
trial, after it happens, it’s also a fact that can be attested. Now, in the General Theory of
Norms, Kelsen proposes a legal theory not only for Law but also for Positive Morals, like the
one based in Christianity.
Positive morals (the kind of system where we are able to recognize the creation and positing
of every moral norm) it’s actually a field where it’s very difficult to establish the boundaries
between positive and natural law. But what Kelsen is proposing is precisely a scientific theory
about it, i.e., a rigid method capable of acknowledging the set of multiple moral norms as a
material capable of rational cognition for a science like normative science. In this sense, it is
important not to mix the theory of its object. But Kelsen is capable of offering a view of
morality compatible with the positivistic approach, making it a fairly possible object of
scientific knowledge:
“Statements about the validity of moral norms are the sentences by which
ethics (understood as a science) describes a certain morality, for instance the
Judaeo-Christian morality posited by Moses and Jesus. In the mouth of God or
of a person considered to be a prophet or the Son of God (i.e. a moral
authority positing moral norms), the word sollen – say in the sentence ‘You
are not to kill’ – has a prescriptive signification. For the sentence is a norm; it
is the meaning of a divine act of will, mediated by the prophet or the Son of
God, and directed to the behavior of human beings. In an ethics presenting –
i.e. describing – Judaeo-Christian morality, the word sollen in the sentence ‘A
human being is not to kill other human beings’ has a descriptive signification.
The sentence of the ethicist is not a norm, for ethics as a science cannot
prescribe anything, but only describe the norms given to it. As a scientist, the
ethicist is no more competent to posit moral norms than the legal scientist is
competent to posit legal norms; the ethicist has no more justification for
67
claiming any moral authority than the legal scientist has for claiming any legal
authority.”71
With this approach, we are trying to show that, concerning at least some aspects, like the
possibility of approach a moral normative system according to a positivist perspective, it is
possible to render the limits between positivism and Natural law at least more flexible. Of
course, there are limits to which this comparison might be pushed. But the positivism qua
method of knowledge of a given set of norms is capable to report also norms involving non-
scientific notions in their formulation, i.e., to attest their validity.
This empirical approach may be even more fitted to “softer” versions of legal positivism, such
as the one proposed by Herbert Hart. The following passage illustrates why:
“The question whether a rule of recognition exists and what its content is, i.e.
what the criteria of validity in any given legal system are, is regarded
throughout this book as an empirical, though complex, question of fact.”72
The problem returns, of course, when those non-scientific notions are no longer simply a part
of the content of the norm, but are also supposed to justify a set of norms as legally binding.
This perspective, defended by the Natural lawyers, cannot be taken seriously according to
positivism, due to the lacking of objectiveness.
So, after this complex context, we seem to be forced to admit that, once in the frame of a
theory aiming the objective grounding and the formal justification of a normative system, the
reason for the validity of a norm, its formal legitimation, cannot make reference to something
simply attested in its content, neither to an ultimate subjective element incapable of being
scientifically approached. Rather than that, the reason that makes the norm binding has to
relate to the authorization of the person enacting it, it has to be about the correctness of this
process of creation. Similarly, the searching for the legitimation will go upwards the normative
chain, until achieving the basic norm, which, in the case of positivism, is to be presupposed as
a fiction. But the fact of being considered as a fiction doesn’t mean that the basic norm is
simply the meaning of an act of thought, as the moral norms according to the Natural law
71 Kelsen, H. (1979, p.154-155). 72 Hart, H. (1961, p.245).
68
approach. In order to avoid that, Kelsen will defend the idea that the basic norm is the
meaning of a non-real act of will, a fictive act of will. We will approach that problematic
question further on our study.73
73 Cf. chapter 6.
69
Chapter 5
Science and Method: the Naturalistic Fallacy
“Science is what we know and philosophy is what we don’t know.”
(Bertrand Russell, 1959)
As we have already seen, the legal theory of Hans Kelsen is marked by a strong interest in
drawing a formal frame of analysis for normative systems. According to this author, the
positive normative systems of Law and Morals, must be neatly separated from the sciences
describing them, i. e., from the legal science in the case of positive law and from the Ethics in
the case of positive moral norms. In the same way, these normative sciences must themselves
be separated from the natural sciences – such as Physics, for example – , inasmuch as the
object of a legal science does not consist in facts expressed by sentences, but instead in valid
norms which are expressed in the form of imperatives or prescriptions.
In this context, Kelsen claims that a common mistake consists in mixing those two separate
domains, the “Is” domain of the facts and the “Ought” domain of the norms, what could give
place to a series of problems in the field of legal theory and legal argumentation, such as the
many quarrels involving the application of logical principles to norms (Jørgensen’s Dilemma
being the most notable example). This misunderstanding is known as “Naturalistic Fallacy”, as
pointed out by G. E. Moore in 1903’ Principia Ethica.
The aim of this chapter will be to explain how Kelsen’s theory relates this fallacy with the
“Principle of Purity”, and how it is possible to better understand and even aim to solve some
classical legal problems derived from the involuntary falling into the Naturalistic Fallacy.
70
5.1 David Hume as a model
Michael Hartney, the translator of the General Theory of Norms from the original German to
the English language tells that, in a personal conversation with Herbert Hart, the last has
confirmed the fact that Kelsen used to say that “Hume was the greatest philosopher of all
times”74. This is a very interesting remark that helps enlightening some of the perspectives
Kelsen later assumes, namely regarding the aims in establishing “solid foundations” for his
scientific theory, as well as the anti-metaphysical approach and the refutation of the notion
of practical reason, together with the general interest in establishing a scrupulous
methodological approach to science.
The most famous quotation regarding this methodological mistake (that was later called the
Naturalistic Fallacy by G. E. Moore) is found in David Hume’s Treatise of Human Nature, first
published at the end of 1738. Hume says:
“In every system of morality, which I have hitherto met with, I have always
remarked that the author proceeds for some time in the ordinary way of
reasoning, and establishes the being of a God, or makes observations
concerning human affairs; when of a sudden I am surprised to find, that
instead of the usual copulations of propositions, is, and is not, I met with no
proposition that is not connected with an ought, or an ought not. This change
is imperceptible; but is, however, of the last consequence.”75
After Hume and Moore, the problems and consequences behind the falling into the
naturalistic fallacy were also expressed by the French mathematician and philosopher of
science Henri Poincaré (1854 – 1912), who also denies the possibility of scientifically qualifying
ethics, as he explains:
“Now, the principles of science and the postulates of geometry are in the
indicative mood and cannot be otherwise. This is also the mood of
74 Also confirmed by Ott, W. (1987, p.539). 75 Hume, D. (2007 [1738], p.244/245).
71
experimental truths; and all the sciences are founded on nothing but the
indicative mood and cannot be founded in anything else. Thus, the most
subtle dialectician can juggle these principles as he wills, combine them, pile
them one on the top of the other: everything he will get from them will be in
the indicative mood. He will never get a sentence which says ‘Do this’ or ‘Do
not do that’, i.e. a sentence which confirms or contradicts morality.”76
More recently, this idea was also developed in detail by the English philosopher Richard Hare
(1919 – 2002), whose approach is based on the notions of frastic and neustic in an attempt to
deal with the problematic notion of practical syllogism, considered the most notable example
of the use of the naturalistic fallacy.
But also Kelsen expressly makes reference to the “Hume Law” in order to corroborate one of
the most basic postulates concerning his theory of legal positivism: that an Ought, a norm,
cannot be derived from an Is, a causal rule or a description of a norm, as those present in legal
science. Thus, to speak about the naturalistic fallacy in the context of Kelsen’s positivist theory
means a recall to be aware of the differences between what is stated in the real actual domain
of our behavior and the evaluation of this behavior (good, bad, just and, regarding the
scientific level, the evaluation of true/false descriptions about the norms) and, on the other
hand, what is prescribed in the domain of Law itself. In the legal domain, the Ought which
characterizes and defines a legal norm cannot be related to those subjective evaluations, but,
instead, it’s objectively determined by the notion of legal validity, which means, as we have
already seen, the specific existence of the norm in a determined legal or moral positive
system.
Therefore, in order to understand the presence of the fallacy in the field of the positive legal
theory, one must understand the existence of this fundamental gap between what belongs to
the factual scope of events and what belongs to the normative scope of norms regulating
those events. In this first attempt to draw the limits between the causal domain of the human
acts ordered by the laws of causality and the legal domain ordered by the laws of normative
imputation, it would be useful to analyze some examples of ordinary acts of the daily life which
may however receive a special legal interpretation. About this issue, it is Kelsen himself who
76 Poincaré, H. (1982/1913, p. 225).
72
gives us several examples of a variety of ordinary acts which can be differently interpreted
according to the Law:
“For example: People assemble in a large room, make speeches, some raise
their hands, others do not – this is the external happening. Its meaning is that
a statute is being passed, that law is created… To give other illustrations: A
man in a robe and speaking from a dais says some words to a man standing
before him, legally this external happening means: a juridical decision was
passed. A merchant writes a letter of a certain content to another merchant
who, in turn, answers with a letter; this means they have concluded a legally
binding contract. Somebody causes the death of somebody else; legally, this
means a murder.”77
To Kelsen, norms are still essentially linked to human actions that are having place in the
factual field of our everyday life. Deryck Beyleveld and Roger Brownsword wisely synthesize
that “To interpret an act in terms of its legal meaning is to view the act in relation to a norm
that is viewed as a legally valid norm”78. But, in order to better understand this relation, we
must analyze more closely this concept of “ought”.
5.1.1 The “Ought”
As we have mentioned before, norms can be taken in consideration when a certain act is being
specifically interpreted: not causally interpreted, but normatively interpreted. This allows us
to say that a norm represents some kind of mechanism of interpretation. More precisely,
according to Kelsen, norms are meanings; they are the objective meaning of a subjective act.
However, the act in question is also somehow restricted: the norm is the meaning of an act of
will from a person, directed to another person’s behavior.
Of course one could argue that all those acts listed in the example abovementioned could also
be interpreted in other different ways, if one would have wanted to do so. The point is,
77 Kelsen, H. (1960, p.2). 78 Beyleveld, D. & Brownsword, R. (1998, p. 116).
73
precisely, that the law gives to those acts a specific interpretation, according to a previous
valid norm existent in a determined system with a corresponding content. Therefore, it is the
norm already belonging to the legal system which gives us the key for interpreting the specific
legal or normative meaning of those arbitrary acts taking place in the causal realm of our
everyday life. Departing from this frame, a new particular norm can then be created, regarding
the very case in question. Thus, this “new” norm will be dependent of two main elements: the
existence of a previous positive valid norm, and an act of will in regard to which this new norm
will be the objective meaning.
Concerning the “objectiveness” in question, Deryck Beyleveld and Roger Brownsword make
this prudent observation:
“As we understand Kelsen, he is not saying (…) that a norm is objectively valid
(that an ‘ought’ has an objective meaning) on condition that everyone agrees
that the behavior prescribed ought to be done. It is not a consensus of wills
that converts a subjective ‘ought’ into an objective ‘ought’. He is saying that
to regard a norm as being objectively valid is (that is, means) to regard what
the norm prescribes as binding upon all whose behavior (actual or potential)
is being addressed by the norm regardless of whether anyone (including the
person who is doing this objective regarding) wishes this behavior or not (that
is, regardless of whether anyone wills this ‘ought’ in the subjective sense).” 79
So, the definition of “norm” is indeed linked to the concept of a command or an order, but
with the important detail that this order must come from an authorized person, as an
expression – mostly in the form of an imperative – of a will. Since this will must come from a
person authorized by the Law itself, there is always a strict relation between the legal
production and the legal power. According to Norberto Bobbio80, to enact a norm is always
to be able to do so, and the authorization comes from no one but the Law itself. This shows
the importance of the fact that the methodological procedures to manage the relations
between the legal norms and handle the variety of possible legal affairs must always come
79Beyleveld & Brownsword, Methodological Syncretism in Kelsen’s Pure Theory of Law. In: Norms and Normativity. Stanley and Bonnie Paulson (Eds.).Clarendon Press Oxford, 1998, p. 118. 80 Bobbio, N. (2008).
74
from the Law itself, something that also illustrates the constant concern to preserve the
principle of “purity”.
So, to sum up, what is needed to create a positive legal norm? First of all, there must be a
prior, more general norm to support the existence of the new one. Then, the “creative”
subject must be an authorized person capable of enacting an objective act of will. Those are
the elements necessary to characterize the norm as a valid one: its specific existence in the
frame of a legal order; the fact of its connection to a legal system via a more general norm,
and the compulsoriness that links up all its addressees. This late obligatory aspect is in fact
the objectiveness aspect involving the act of will coming from the Judge or the Legislator. An
act of will with a subjective meaning consists only in a command, without legal force, it means
the expression of a personal will towards a specific case, and doesn’t even enters the
normative legal level. Instead, the act of will with an objective meaning has to come only from
an authorized, impartial, neutral person, and it’s precisely the meaning of this objective act of
will that represent a legal norm.
In this perspective, we can sustain that, while the command is the expression of a desire, the
norm is the expression of a duty, of an “ought” (Sollen). The notion of “Ought” introduces us
to the separation between the legal prescriptive field of the norms – the domain of the
“Ought” – and the factual descriptive field of the legal science – the domain of the “is”. This
distinction addresses us again to the main problem we intend to analyze here: the Naturalistic
Fallacy.
5.1.2 The “Is”
Now, in Kelsen’s theory, it is very clear that the principle of “purity” guiding the work of the
jurist is directly linked with the problem of the naturalistic fallacy. The fact that a norm can
only be created via the existence of a previous norm, or, said in another words, that an ‘ought’
can only be created once given the existence of a previous ‘ought’, states the necessity of
staying in one and the same scope, namely, the normative, legal one. The “purity” here also
indicates that it is necessary to stay inside this normative level in order to create new norms,
and that the “ought” characterizing the norm must not be mixed with the factual elements
pertaining to the domain of the “Is”. Practically speaking, this means that the norm must not
be confounded with the external elements which may be inserted in this legal level. One
75
consequence would be the impressive remark that norms don’t need to express a just content
in order to be accepted as valid norms. Fairness or justice are outer elements depending on
subjective moral principles, and stay completely strange to the “pure” field of legal positivism.
The same might happen with psychological, sociological and political aspects involving norms.
Kelsen argues that the admission of such elements will lead to a fallacy, that one must not
deduce that something might be the case simply because something is the case. Similarly, one
cannot establish that something might be the case simply based on the fact that something is
(subjectively) believed to be the case. One field is regulated by reasoning processes, the other
by intentional acts of will.
So, the naturalistic fallacy appears in Kelsen’s theory when is question of separating the
normative field from the indicative one. In a practical perspective, this is represented by the
division or the gap which needs to exist between the system composed by norms on one side,
and the legal science describing those norms on the other.
We must note that the great distinction between the legal norm and the sentences of the
legal theory (or legal science) lies in the fact that the sentence can, among other qualifications,
also be qualified according to truth values, while the norm can only be said to be valid.
Actually, even the facts themselves linguistically expressed in the sentences cannot be valued
as true or false, but only as existing or not existing: true or false are only the descriptions made
about the existence of those facts. The predicates true/false can be said only in relation to
elements of the “is” domain, to the descriptions of the facts which can be made by any
science. But they are not linked to the very object of these sciences, as the natural facts, for
instance. On the other side, the norm is exactly the object being described by the legal science,
and it can, equally, only be said to be existent or not existent. The point is that when the norm
is said to be existent, this means that it is a valid norm (here “valid norm” is actually a
pleonasm), and the very existence of the norm in the legal system constitutes already its
validity. The validity of a norm is therefore simply and nothing more than its specific existence.
5.2 One Example of the Fallacy: Jørgensen’s Dilemma
According to the traditional conceptions of logic entailment, which Kelsen searches in the
traditional notion of Aristotelian syllogism, the conclusion will follows from the two premises
if and only if the conclusion cannot be false if the premises are true. This is the “rule of
76
inference” Kelsen analyses when dealing with the question whether norms can be created via
a similar form. In the case of the syllogism, the inference is logically valid or not depending on
the evaluation of the sentences according to the values of true and false. The problem arises
when imperatives, commands, legal or moral norms are inserted in this kind of syllogism. This
happens in the so-called practical syllogisms, where a norm is supposed to be “created” from
a general norm, together with the fact attesting its content. But, clearly, since norms cannot
be evaluated according to truth values, they cannot play the role of premises and/or
conclusion, because there’s no way to evaluate the validity of a norm only by means of logical
reasoning.
All of that appears to be straightforward. Nevertheless, when we first start looking to some of
the examples of these so-called practical syllogisms, the conclusion seems to follow as
inescapably as in a valid legal argument expressed by a theoretical syllogism. Let us take the
example:
Love your neighbors.
John is your neighbor.
Love John.
It seems impossible to argue against the validity of such an argument. This difficulty was first
expressed by Jørgen Jørgensen in 1937, and was named the “Jørgensen’s Dilemma” by Alf
Ross, in 1941. Jørgensen explains himself:
“So we have the following puzzle: According to a generally accepted definition
of logical inference only sentences which are capable of being true or false
can function as premises or conclusions in an inference; nevertheless it seems
evident that a conclusion in the imperative mood may be drawn from two
premises one of which or both of which are in the imperative mood.”81
As we’ve seen, according to Kelsen, there is a “methodological abyss”, a gap between the
“Ought” (Sollen) domain, tenanted by norms; and the “Is” (Sein) domain, tenanted by
sentences. Jørgensen’s dilemma puts in question which kinds of interaction seem to prevail in
81 Jørgensen, J. (1937, p.290).
77
the relation between prescriptions or imperatives (Sollsätze) and descriptions or sentences
(Sätze).
But, when considering the differences between the norms and description of norms, Kelsen
keeps insisting that the two cannot be reducible or even comparable. One is the Law, and the
other is the science that describes the Law as its object. Kelsen says:
“‘Is’ and ‘Ought’ are purely formal concepts, two forms or modes which can
assume any content whatsoever, but which must assume some content in
order to be significant. It is something which is, and it is something which
ought to be. But no specific content follows from the form.”82
In order to explain the relation between form and content regarding the legal norm and the
legal proposition of the science, Kelsen introduces the notion of modally indifferent
substrate83. So, if we consider the following sentences:
1. “Al Capone ought to pay the taxes.” and
2. “Al Capone pays/had paid his taxes.”
we have one modally indifferent substrate, which is te behavior “paying the taxes”, which
happens to be under two different modes, namely, once under the “ought” form in 1. and
after under the “is” form in 2. Their content is identical, but the linguistic formulations
themselves are not comparable, correspondent or deductible one from another.
So, concerning Jørgensen’s dilemma, if we stick to Kelsen’s conception of the modally
indifferent substrate, we must notice that, in fact, the modally indifferent substrate is neither
true nor false in itself. It can only be the content of a true/false sentence, in the same way
that it can only be the content of a valid norm. But there is no intrinsic correspondence
between a true/false sentence and a valid norm; they only happen to have a similar modally
indifferent substrate.
82 Kelsen, H. (1979, p.60). 83 The notion of modally indifferent substrate and its relation to Jørgensen’s dilemma will be the main topic of the chapters 8 and 9.
78
So, when Jørgensen says that the answer to the dilemma is only a matter of translation (true
indicative sentences could be identified and “translated” into valid norms or imperatives),
actually there is absolutely no correspondence between an imperative or a norm and an
indicative sentence: the fact is that we are simply dealing with a modally indifferent substrate,
which is presented or “dressed up” in two different modes. To sum up, the point of interest is
that Kelsen’s notion of the modally indifferent substrate is outside the bounds of any bivalent
logical approach.
To Kelsen, every norm must have content, the “modally indifferent substrate”, which is not
the indicative factor, as Jørgensen’s believed the content of the imperative to be, but rather
the very behavior wanted by the imperator. So, in one way, the person enacting the norm
knows the content of the norm that will be enacted, his will will be concerning some specific
behavior, that will be expressed or indicated by the norm to be created. This act of thought
concerning the norm is somehow the “first” step preceding the act of will of the agent. And
on the other way, also the addressee of the norm must likewise understand what is he
supposed to do, how is he supposed to behave, and he does that by observing what is the
content of the norm. Let’s see how this relation is manifested in an example. Consider the
following situation:
“Every citizen should pay the taxes” consist in a valid norm, let us say. This norms also applies
to, for example, the citizen Al Capone. Once a year this individual receives a letter saying how
much he has to pay to the government, as income taxes. The legal norm behind the letter,
putted in a simpler way, says that “Every citizen must pay his/her taxes”, even if this very
prescriptive element is not explicitly formulated in the text addressed to Al Capone. It is
somehow implicit on the fact of having the bill addressed to him. The norm-positing subject
can nevertheless justify his demanding for the money, if needed, by enumerating the benefits
to be seized by the citizens, like the improvement of social and public services, the reduction
of budget deficits (though he doesn’t need to, given his authorization). In Kelsen’s theory,
those aspects related to having a norm posed to one’s behavior and immediately knowing
how to proceed are to be seen as acts of thinking regarding the norm, and they have nothing
to do with the willing positing the norm as valid and binding. They have nothing to do with the
modally indifferent substrate, which is, in this case, the neutrals “to pay the taxes” or “paying
the taxes”. Those neutrals are the content of the norm. Thus, by receiving the letter and
checking the sum to be paid, Al Capone knows the implicatures of is adhering or ignoring the
79
bill. He knows everything that’s “behind” the bill, by “dressing” the normative content in a
descriptive “outfit”, by means of his acts of thought.
So, what happens when we are confronted with the notion of practical syllogism is that we
are confounding our act of thought about the content of the norm and the act of will providing
the norms’ validity. The general problem of the relation between imperatives and statements
is much simpler when we only take into consideration the legal field (apart from all the other
imperatives or “ordinary” orders). Then we can argue that, even if we somehow know the
content of the norm which is “supposed to be created” by the Judge in a simple case, it’s only
the enaction of the individual norm by the Judge that will render the imperative a legitimate
valid legal norm.
The misleading “practical syllogism” actually works in our daily lives because we are in fact
simply reasoning about the norms involved in the inference, through acts of thought. The
fallacy emerges when we try to trespass this level and admit that we are actually creating a
valid norm from the existence of a true sentence plus the fact of the existence of a previous
norm. It is, indeed, natural that we try to understand what the norm is demanding, in order
to know how to behave. The minor premise which poses a fact is actually the context in which
we legitimate are reasoning by acts of thought, and the major premise (indicating a norm in
the practical syllogism) poses the norm that we must understand, comprehend, which
presents the behavior wanted as its content. If I understand the first premise in the context
of the second premise, I’m able to know which behavior I’m supposed to have, I can tell what
the modally indifferent substrate is. All this procedure does not touch the normative level,
because no objective act of will has yet being considered or enacted. It’s all a question of
knowing what is at stake, i.e. of considering the given facts through mental processes.
Moreover, we must also notice that, as the imperator, it is perfectly possible to understand
the content of a general norm, together with the corresponding fact and still do not want to
posit the individual behavior as obligatory in the conclusion. The particular norm posed as a
conclusion in the practical syllogism will only be a valid binding norm after the adding of an
objective act of will. Until that moment, we remain in the realm of the Is, where no valid legal
norm might be created.
5.3 Final remarks on this topic
80
The legacy of the naturalistic fallacy is relevant to make the important remark that norms
cannot be the result of conclusions of our reason or, said in another manner, of rational
deduction. David Hume states that:
“Reason is the discovery of truth or falsehood. Truth or falsehood consists in
an agreement or disagreement either to the real relations of ideas, or to real
existence and matter of fact. Whatever, therefore, is not susceptible of this
agreement or disagreement, is incapable of being true or false, and can never
be an object or our reason”84.
As we have seen, Jørgensen’s dilemma is a perfect example of the application of this
misleading procedure.
However, Kelsen’s theory represents a fine example of the application of David Hume’s
recommendations warning us to the mistakes derived from this mixing of the two different
levels, Is and Ought. In the frame of our study, this represents an appeal from legal theory to
the methodological principles coming from the philosophical domain. The interesting detail
to notice is that the notion of naturalistic fallacy displays, in a perfect manner, Kelsen’s
concerns about the correct treatment of the normative field as the object of the scientific
approach.
The positive outcome of the avoiding of the naturalistic fallacy is the widening of the
perspective about the legitimate approach of norms and the process of normative creation.
In this sense, by respecting the boundaries of prescriptive and descriptive fields, one can
clearly understand that it is perfectly acceptable to make inferences about the fact of the
existence of a norm, in order, for example, to justify the legitimation of the normative creation
once it has already happened. This is allowed by the important notion of modally indifferent
substrate, which we briefly introduced in this chapter, but that will be further treated in
another part of the present study.85 By attempting to the fallacy, we can fully grasp the idea
that norms have form and content in the same manner as the descriptions about their validity,
but then we must also recognize that, nevertheless, this aspect does not allow the possibility
84 Hume, D. (2007 [1738], p. 240). 85 Cf. chapters 8 and 9.
81
of deriving norms from facts or descriptions, nor to establish direct logical relations between
them.
The fact that norms are supposed to be left in this “pure” prescriptive level does not attest
that they are unreachable or inaccessible to human discourse, evaluation or study. It only
attests that those aspects, even if fundamental for the juridical scientific approach, are still
irrelevant for the creation and application of legal norms.
82
Chapter 6
The Non-Existence in Legal Science
“It's no wonder that truth is stranger than fiction.
Fiction has to make sense.”
Mark Twain
If the presence of fiction in natural sciences is sufficiently known and accepted, the same
doesn’t seem to be the case when it comes to its legitimacy in legal science, as we will see in
the present chapter.
The presence of fictions as a part of the legal system, within the formulation of the norms, can
actually be traced since Roman law, even if its legitimacy also in that level remains a matter
of great divergence among critics. Nevertheless, the legitimacy of the use of fictions by natural
sciences or philosophy is attested by famous examples of thought experiments, for instance.
It is sufficient to think, for example, about Schrödinger's cat or the twin paradox (or Langevin
paradox) in physics, as well as the twin Earth thought experiment in philosophy. They function
as a scientific tool allowing to more easily explain and understand the concepts or theses they
are referring to.
Considering this context, in this chapter we will analyze the use of fictions made by this special
kind of science dealing with the regulation of our behavior, namely legal science. The
fundamental question to be approached is whether the basic norm, to be presented as a
fiction, has a normative or prescriptive profile.
Our aim is to analyze the use of fiction by the legal science under the light of the legal theory
proposed by Kelsen, especially concerning his proposal that the legitimization of the whole
positive legal system is based on a fiction, called the Basic Norm (Grundnorm). The thesis we
83
sustain is that this “norm” must be seen as a purely methodological or scientific tool (just like
the thought experiments of physics and philosophy), and not as an ordinary legal norm among
others in the positive normative system. We will try to elucidate how can a fiction display such
an important scientific function and still preserve the “principle of purity” of the Kelsenian
legal theory.
6.1 Legal science: Meaning and Particularities
As we have already many times insisted in express, it is a central aspect of the kelsenian legal
philosophy the fact that the Law, namely the overall set of norms, must be clearly separated
from the science that describes and studies this normative set. Another differentiation
cherished by Kelsen separates neatly this kind of normative science, which takes the legal or
moral norm as its object, from the natural sciences, which takes the facts of nature as its
object. These two ideas constitute the methodological frame in which Kelsen formulates his
Pure Theory of Law. It is also the notion of “purity” that allows one to deal with each one of
these Kelsenian paradigms.
The fact of confusing Law and Legal Science was a common mistake made by some of Kelsen’s
contemporaries86. Certainly, one should concede that the content of legal science, in the
context of legal Positivism, can indeed sometimes be very confusing. The theory of Law
describes the norms of a particular normative system87, but the point is that the norm
described by the science has no longer its normative power or, in the terms of Kelsen, it is no
longer a valid norm. It is exactly the fact of the validity of the norm that is being described,
and a fact cannot (legally) compel anyone.
But let us start from the beginning, by recalling some aspects concerning the term “legal
norm” according to Kelsen’s positivist theory.
The definition of “norm” is relied to the concept of a command or an order, with the important
detail that this order must come from an authorized person, as an expression – mostly in the
form of an imperative – of a will. Since this will must come from a person authorized by the
86 This “confusion” is linked to the denial of the duality of “Is” and “Ought”. In order to see how Kelsen treats and analyses several classical cases, Cf. KELSEN, H (1979, p. 63-82). 87 Cf. Spaak, T. (2005, p. 397- 414).
84
Law itself, there is always a strict relation between the legal production and the legal power88.
The norm is the meaning of an objective act of will, coming from an authorized person. It is
marked by the presence of the “ought” particle89, meaning that we’re not in the domain of
the “Is”, but in the domain of the “Ought” (Sollen). Kelsen says:
“When someone commands or prescribes, he wills that something ought to
happen. The Ought – the norm – is the meaning of a willing or act of will, and
– if the norm is a prescription or command – it is the meaning of an act
directed to the behavior of another person, an act whose meaning is that
another person (or persons) is to behave in a certain way”.90
According to Bobbio91, to enact a norm is always to be able to do so, and the authorization
comes from no one but the Law itself. This shows the importance of the fact that the
procedures to manage the legal norms and handle the legal affairs must always come from
the Law itself, according also to the principle of “purity”92.
6.1.1 Some examples
So, again, what is needed to create a positive legal norm? First of all, legal validity is linked to
the specific existence of a norm; second, to the fact of its belonging to a legal system; and
third, to the compulsoriness that links up all its addressees. Those three elements are
nevertheless related in an intrinsic manner. In our context, this late obligatory aspect is very
important, and it is in fact the objectiveness of the meaning given to the act of will coming
from the Judge/Legislator. A subjective meaning of the act of will consists in a command
without legal force, it means the expression of a personal will towards a specific case, and
that’s not what Law is about. The objective meaning of the act of will comes only from an
88 Cf. van Roermund, B. (2000, p. 201-222). 89 Even if the norm in question is not mandatory, but concerning permissions, empowerment or derogation, the “ought” element is always preserved. 90 Kelsen, H. (1979, p.2). 91 Bobbio, N. (1997). 92 What is implicit in this conception is that there is no place to logical treatment in the inner domain of Law, i. e. in the production or interpretation (decision) of legal norms.
85
authorized, impartial, neutral person, and this objective meaning of this act of will is the legal
norm93.
Now let us think of one example. Consider a “community” of pirates, for instance. Their
community is hierarchically organized, with solid efficiently rules of behavior, which are
enforced by the systematized application of sanctions. They seem to experience a very solid
and efficient legal order! Nevertheless, regarding this example, Kelsen would certainly point
out the lacking of objectiveness concerning the meaning of the act of will regarding the
pirates. In the regress for searching the justification of those norms, there is no grounding for
the enaction of the multiple norms composing their normative order, which might be rather
seen as an (organized) agglomerate of orders94.
While the command or the order is the expression of a desire, the valid norm is the expression
of a duty, of an “ought” (Sollen). The notion of “ought” introduces us one more time to the
separation between the legal prescriptive field of the norms – the domain of the “ought” –
and the factual descriptive field of the legal science – the domain of the “is”.
Concerning the relations between the science and its scientific object – here, legal science and
Law itself – the English legal philosopher Herbert Hart95 uses a great example to illustrate the
fact that, even if the legal science deals with valid norms, it doesn’t have any normative legal
power96. He uses the example of the relation between someone who speaks a foreign
language and his, let’s say, English interpreter. If a German captain in a concentration camp
says out loud to his English or American prisoners “Stehen Sie auf!”, the interpreter will
probably also say out loud the words “Stand up!”. The interpreter will do his best to show, by
his intonation maybe, or by the expression in his face, that what the Captain said was not a
begging or a simple request: it was an order. The point is: how do we must consider the
sentence “Stand up!” in respect of its original in German? Is it a second order? Is it the same
order? Is it the emission of an order? Well, the interpreter has no authority to emit orders. His
93 By this approach, an objective command is not only the psychic event of the expression of a will. This can be seen in the case of a testament, for instance. In a valid testament, the subjective act of will of the person in question obtains its objectivity trough the Law: once it is legally legitimated, the command of the person in question will remain beyond his own existence, when he will no longer be able to express his will. This demonstrates the independency of the compulsoriness of the command from the subjective act of will. 94 Actually, what turns a legal norm into a binding norm is the fact of assuming the existence of the Basic Norm in relation to such a normative system. We come to this issue later in this text. 95 Hart, H. (2005 [1963]). 96 This aspect is essential to our future analysis of the Basic Norm as an element of legal science, and not of Law itself.
86
job is to interpreter the orders of the captain and, if the order is obeyed or not, it was the
Captain who was therefore obeyed or disobeyed. Hart then says that, in perfect accordance
to Kelsen’s theory, what happens in such a case is that the captain’s order is being described
by the interpreter, that the “Stand up!” was an order in a descriptive sense, that the original
imperative was used in a descriptive, not prescriptive, sense.
While the example of the pirates show the absolute necessity of a fundament that gives the
unity of a legal system, allowing it to be rationally acknowledged and scientifically
investigated, the example of the interpreter displays the relations between those norms with
what is produced by the legal scientist.
6.1.2 Dichotomies in Kelsen’s Theory
As we already could see, the legal philosophy of Hans Kelsen is essentially marked by its
manifold dichotomies. When questions about the purposes of the legal sciences are at stake,
another dichotomy then arises: the legal science is a descriptive science inasmuch as it doesn’t
make any prescription; but, at the same time, the objects of its descriptions are not
statements, but prescriptions. Consequently, the normative science makes descriptions about
prescriptions. Norberto Bobbio finely analyzes this issue:
“‘Normative’ is in opposition (already in the Hauptprobleme) not to
‘descriptive’, but to ‘explicative’; and, in parallel, ‘descriptive’ is in opposition
(especially in the last works) not to ‘normative’, but to ‘prescriptive’. Given
that the doubles ‘normative-explicative’ and ‘prescriptive-descriptive’ do not
superimpose themselves, there is no contradiction in affirming, as Kelsen
does, that the legal science is at the same time descriptive and normative:
descriptive in the sense that it does not prescribe, normative in the sense that
the things being described are not facts, but norms, i.e., it is descriptive not
about what exists, but about what ought to be. As Sollsätze, the propositions
which characterize the legal science are distinguished in one hand from the
87
Seinsätze belonging to social sciences (causal), and, on the other hand, from
the Sollnormen of any normative system.”97
What Bobbio tries to explain is that the normative science, despite the fact of dealing with
norms – where the “normative” term comes from –, does not make use of a prescriptive
discourse, that is, its descriptions don’t have the aim of changing the behavior of others. They
are placed in the domain of the “is”.
6.2 The Basic Norm as a scientific fiction
After the preceding considerations, one should be tended to imagine what the legitimate
object of a legal science really looks like. Kelsen sustains a hierarchical vision of the legal
system (the object of the legal science), in the form of a triangle or a pyramid, where the base
is formed by the particular norms created98 by the Judge in the tribunals. Those norms must
be based on the validity of more general norms, and that searching for grounding happens
until we arrive at the Constitution of a Country, for example. Then, one could still regress and
go up on the pyramid to attain the very first Constitution of a Country, historically speaking.
This searching is performed by the legal science, in its analysis of the legal norm as an object
capable of being rationally described as a part of science.
The legal positivism (contrarily to the “antagonist” schools of thought such as the legal realism
or the Natural law tradition) focuses on the claim that the norm and the Law are human
constructions – they must be posited, enacted by someone – and the central notion is not the
efficacy of a norm, or the moral value of a norm, but its validity, its existence and legitimate
belonging to the legal system. The scientific justification of the validity of a system, of its
legitimate existence as such, is based on the validity of the multiple norms composing it.
97 Bobbio, N. (2008, p. 58). This quotation was a personal translation of a Portuguese version of this book. 98 Maybe it would be more correctly said that the Judge doesn’t create any norm, he only applies the norms present on the system. But it won’t be entirely wrong to say that particular or individual norms are actually created by the Judge, since they ultimately result from his particular act of will.
88
6.2.1 The Searching for Justification
A special question that arises in the justification of the normative system refers to what would
make a unity from the multiple norms of a Country, for example. Well, the first response to
that question would probably mention the Constitution. The particular norms applied by the
Judge (by his particular acts of will) in the tribunals would be supported by the general norms
created by the Legislator (by his particular acts of will), and that’s how we arrive, at the end,
at the Constitution. But here we have to face a problem: from where does the Constitution
obtain its validity? From a prior Constitution, one could say. But from where does the first
Constitution of a Country (that earlier here we have placed in the top of the normative
“pyramid”) obtain its own validity? Or, in other words, the person who enacted the first
Constitution of a Country, the very first Legislator, was authorized by whom to perform such
an activity? How did it get legal legitimacy?
Here’s the difficulty: every time we’ll put a prior legal document as posited in this regression,
the question about its legitimacy will be at stake. To solve this problem, Kelsen will say that
what gives the foundations of the legal system cannot be a positive norm, a written document,
because this norm would forcedly again have to be supported by a previous norm, and the
person enacting this norm would have to obtain the power to create it from anyone from a
even higher degree of competence. So, Kelsen will say, what legitimizes the creation of the
first Constitution of a Country or, more generally, what legitimizes and gives the unity to a
whole legal system, is in fact not a positive, but a fictive norm, called the Basic Norm99.
What we will advocate here is that this Basic Norm is not placed in the legal system, on the
top of the pyramid, as many critics of Kelsen suggest. We will defend that the Basic Norm is
nothing more than a scientific fiction, a methodological tool. But, more precisely, our original
approach will suggest that the regression in the seeking for the legitimation of the system
must actually stop at the Constitution, and what we actually need after arriving there is not a
new element, but a scientific tool used to conveniently consider this last object according to
the positivist claims. It’s only when we arrive at the top of the pyramid that we finally need
99 If we were to (wrongly) give some formulation to the Basic Norm, for didactic purposes let’s say, it would sound like “We should do as the first Constitution of this Country tells us to do”.
89
the notion of Basic Norm, in order to rationally consider the first Constitution of a Country as
the elementary component of the legal order.
In fact, we believe that the legal system should be illustrated by an isosceles trapezoid instead of a triangle or pyramid.
The fact of clearly placing the Basic Norm in the scientific level will bring many clarifications
to the kelsenian approach. First, the principle of purity will not be harmed by the presence of
a fiction. Second, the theoretical aim of the Basic Norm will be preserved, since it’s clear that
the Basic Norm is not a norm to be respected by the individuals – it is not an effective norm.
Lastly, the only problem that could remain, and also the more difficult one, will be related to
questions of the form: “But why do Kelsen call the Basic Norm a norm?”.
Our answer to this question makes reference to the fact that the fictive character of the Basic
Norm concerns precisely its validity (and, as far as we remember, “validity” equals to “specific
existence of a norm”), so it is the validity of this “norm” that makes reference a fiction. But
what is fictional about the Basic Norm?
6.2.2 Understanding the Fiction
90
There is a difference between the fiction of the Basic Norm and the usual fictions present in
Law among the other norms of the legal system100. The most common types of legal fiction
will include contradictory elements expressly in the normative formulation, leading to an
apparently impossible situation, with the objective of attaining a specific goal. Let’s see some
examples:
“In property law, a husband and wife could be treated as one person. In family
law, a child’s will is attributed to his guardian. In the interpretation of wills,
one spouse may be deemed to have predeceased the other, even though that
may not in fact have been the case. Under the attractive nuisance doctrine, a
child who trespasses is treated as having been invited onto the defendant’s
land. In immigration law, an alien may be considered to be legally excluded
from the United States even though he is physically within its borders. In civil
forfeiture proceedings, property itself may be named as a party to litigation
(...).”101
Clearly, the Basic Norm doesn’t share the same characteristics of the so-called legal fictions,
that is to say the norms including contradictory elements in their formulation. So, where is
the fiction concerning the Basic Norm? How can we recognize it as a fiction?
6.2.3 A Little Help of Philosophy: Vaihinger’s “As-If” approach
To elaborate the notion of Basic Norm as a fiction (a conception which is present only in the
last writings of Kelsen – before the Basic Norm was seen as an hypothesis102), Kelsen searches
for the foundations of his conception of fiction in the work “The Philosophy of As-If”103, by the
German philosopher Hans Vaihinger (1852-1933). To Kelsen, the fictional element concerns
100 For the relations between fictions and the normative level, Cf. Sievers, J. & Rahman, S. (2011). 101 Student Author (2002, p. 2233-2234). 102 In his previous writings (Kelsen, H. 1960), Kelsen tended to see the Basic Norm not as a fiction, but as a hypothesis. This approach was abandoned because Kelsen got aware from the fact that the “existence” of the Basic Norm would never be able to be verified. Contrary to hypotheses which are constructed with the very aim to be lately confirmed or falsified, the Basic Norm is presupposed with the consciousness of its impossibility of ever being verified. 103 Vaihinger, H. (1965 [1911]).
91
the act of conceiving the foundations for the legal system: we make “as if” there were a higher
norm above the first Constitution, in order to stop the searching for legitimacy and to give to
the system a unity.
So, in the case of the Basic Norm, the fictive element cannot be placed in the formulation of
the norm. To formulate the Basic Norm expressly in the legal system means to submit it to a
higher justification, is to presuppose that there is a higher power above it, in order to
legitimize its creation. So, it is not the case that its formulation will give rise to contradictions
but, more fundamental than that, the very fact of being formulated will already abort its
specific function. Once formulated, the Basic Norm is legally enacted and enters in the circular
searching for its own legitimacy. Kelsen himself explains this difference, showing that the fact
of accepting the necessity of a fictive Basic Norm does not imply accepting the presence of
fictions in Law:
“According to Vaihinger, a fiction is a cognitive device used when one is unable
to attain one’s cognitive goal with the material at hand (1935:13). The
cognitive goal of the Basic Norm is to ground the validity of the norms forming
a positive moral or legal order, that is, to interpret the subjective meaning of
the norm-positing acts as their objective meaning (i.e. as valid norms) and to
interpret the relevant acts as norm-positing acts. This goal can be attained
only by means of a fiction. It should be noted that the Basic Norm is not a
hypothesis in the sense of Vaihinger’s philosophy of As-If – as I myself have
sometimes characterized it – but a fiction. A fiction differs from a hypothesis
in that it is accompanied – or ought to be accompanied – by the awareness
that reality does not agree with it.” 104
6.3 Where is the Basic Norm?
If we cannot formulate the Basic Norm in the terms of a prescription, it clearly means that it
is not a regular norm, a valid norm or an efficient norm. We cannot obey or disrespect the
104 Kelsen, H. (1979, p.256).
92
Basic Norm. Its purpose is not that of controlling our behavior. That is precisely the key to
understand the notion of Basic Norm in Kelsen’s theory. We sustain that the searching for the
justification stops at the First Constitution: whatever comes after concerns no longer the Law,
but only the legal science. By that perspective, the Basic Norm has no interest to the lawyer
or the judge, neither to the defendant or the complainant. It is a matter of scientific
understanding of the system, in the terms that the legal system could not be rationally
described without the notion of the Basic Norm. The Basic Norm appears in the act of the
jurist when he is confronted to the problem of the formal justification of the whole system of
legal norms. The jurist need to presuppose the existence of a fiction of the Basic Norm,
otherwise the legal system would be nothing but a chaotic heap of norms.
Our point is to clear up the fact that the Basic Norm finds its fictive element in the act of doing
“as if” it existed. To conceive the Basic Norm consists in an act that involves the presumption
of its existence, even though we know that the Basic Norm cannot exist. The fictiveness of this
“norm” opposes anyone to identify it in any level: the Basic Norm is not a positive norm, but
neither a formulated principle found in legal science. The point is that nobody is able to
formulate, to construct or linguistically express such a thing as the Basic Norm. If it is enacted,
we enter in the vicious circle and we need a new element to justify its existence as a norm. If
it is formulated as a scientifically principle, it loses its purpose, which is to end the regression
for justification and serve as the base for the legal system. What we defend is that the key to
understand the needing of a Basic Norm is to see it as a part of the process of justifying the
system, as a scientific attitude in front of the legal structure. To put it in another way: we only
can understand the legal system as a rational ensemble of norms from the moment when we
presuppose the fiction of a Basic Norm as being the starting point of the normative creation.
The presupposition of the Basic norm is the condition sine qua non for the legal system to be
a valid one (as a whole).
That’s why do we have to call the Basic Norm a norm. The fiction about the Basic Norm is
found precisely in its validity. We have to make “as if” there was a Basic Norm, because we
know that it cannot be valid, it cannot legally exist. But we make “as if” it could give the
legislator the power to enact the first Constitution, to turn the Constitution into a binding
document.105
105 We must remember that the first aim in using a fiction in science is always its practical utility. Loewenberg, J. (1912, p.717) explains clearly the notion of fiction according to Vaihinger’s theory: “Fictions, in Vaihinger’s usage, are not identical with figments, such as centaur or fairy, nor are they
93
This approach can answer a famous critic to the notion of Basic Norm: that the whole positivist
legal system will finally be built over a fiction. We defend here that the whole dynamic
structure of legal power related to the normative creation is in fact founded on a fiction, but
not the legal normative system, namely, the set of legal norms themself. The searching for the
justification of the validity of the norm can be retraced until the first Constitution of the
Country, and the process stops there. Materially, there is no legal element before the first
Constitution – since the Basic Norm has no legal validity. We need the concept of Basic Norm
only in a scientific level, as a presupposition in the mind of the jurist, as an intentional act
towards the object of study, in order to give the system its unity and its legitimacy as a whole.
Also in philosophy, the topic of non-existence has a solid and long tradition, especially in the
field of analytical philosophy. Even if the notion of existence in philosophy is not the same as
legal existence of validity, the fiction is also considered, in the philosophical domain, as a
legitimate part of a theory and a useful scientific or epistemic device (also attested by the
various and sometimes famous examples of thought experiments – which are also “as-if” kinds
of construction – in philosophy). Surely we could also analyze those approaches to the
question of non-existence, but none of them seem to really illustrate Kelsen’s intentions with
a notion such as the basic norm to the same extent that the philosophy of “as-if” proposed by
Hans Vaihinger. It is, however, disappointing to notice the almost complete disregard to the
writings of Vaihinger by the nowadays scientific and philosophical community. Nevertheless,
this fact might be explained, even if it’s not justified, by the original and unorthodox
interpretations that Vaihinger has made concerning the works of Kant, as well as the severe
skepticism derived from his view concerning fictions as being the only means to human
epistemic access to the world.
So, going back to our topic and answering to the question from this section, the Basic Norm is
definitely not a part of the legal system, but it is neither a part of the legal science. It is a
presupposition from the jurist, precisely in the moment of his scientific consideration of the
normative legal system as its object of study, being such a presupposition prior to any possible
consideration about a specific legal system.
hypotheses capable of verification. They are deliberate devices (Kunstgriffe) on the part of thought for the practical purpose of successful orientation in and perfect control over the environment. Theoretically they are absolutely valueless. Applied with a knowledge of their fictitious character, they will lead to the intended practical results.”
94
That will mean that, without conceiving the notion of Basic norm, without presupposing it in
such a manner, no legal system can be rationally analyzed, studied or interpreted by a science.
Without the fiction of a Basic Norm, the legal system will be just a multiple set of norms
without inner cohesion, without beginning or end. The Basic Norm, according to our approach,
should be seen as an inherent methodological procedure, prior to any consideration of a
positive legal system.
6.4 Final Remarks on this Topic
The fiction of the Basic Norm has to be clearly separated from the cases of legal fictions that
we know since Roman law. The Basic Norm has no legal validity, has no regulating power in
relation to our acts, it cannot even be formulated. In order to dissolve the confusion saying
that Kelsen loses track in inserting a non-positive norm in its legal theory, we must understand
that the Basic Norm is nothing more than a scientific methodological device, to be used only
by the scientist. It has no direct relation with the ordinary norms of a system, it only works as
a presupposition from the part of the jurist when it’s question of considering a specific order
as a whole valid and legitimate legal order.
In this perspective, it is legitimate to consider the Basic Norm as a tool or a device that allows
the scientist to approach the legal system and, if we must make a metaphor here, we could
borrow the beautiful image of Kant and say that the legal system can only be rationally
regarded through the lenses of the Basic Norm.
The fictive character of the norm is explained by Kelsen’s fascination with the work of Hans
Vaihinger. Vaihinger’s “philosophy of as-if” is perfectly adapted to what Kelsen wanted for the
Basic Norm: to show that the use of a fiction in the frame of a science is perfectly viable in
order to attain an objective that could not be attained otherwise: to give to the legal system
the unity necessary to the legal analysis. To completely understand the necessity of the Basic
Norm is to accept that it has to be completely separated from the legal system; it means to
not be misled by the “norm” in Basic Norm.
When Kelsen adopted Vaihinger’s “philosophy of as-if”, he was certainly attracted by this
notion of “making as-if” something were the case, and what we tried to clarify in this chapter
95
is the emphasis on the “act” of doing as if the Basic Norm existed. The Basic Norm is something
that we use, it is a scientific tool. It is not a legal element and, if we stay strict, not even a
methodological element, in the sense that it doesn’t really make part of the legal science, but
it must be a part, and it is an essential element, of the scientist’s approach towards its objects,
the legal norms. Without the presupposition of a Basic Norm by the jurist no theory can be
founded, and no science can exist.
96
Chapter 7
Normative Conflicts and Temporality in Law
“Quod principi placuit legis habet vigorem.” Ulpianus
Normally, the main context giving rise to questions concerning the temporal aspect in Law are
the ones involving cases of conflicting norms. To Kelsen, situations of norms in conflict must
be solved not via the application of logical principles, namely by the principle of non-
contradiction, and not even by traditional legal principles such as lex posterior derogat priori,
but rather by the application of a special kind of legal norm itself, to be called derogating
norm.
The purpose of this chapter is to give an overall view of how norms relate to the temporal
aspect in the frame of the kelsenian positivistic theory, namely how a norm “exists” in time,
and how it can lose its validity over time. We will also analyze the question of retroactivity,
and the question of the lacking of efficacy after a determined period of time and how it affects
the losing of legal validity. Lastly, we aim to investigate how fictions, such as the Basic Norm,
relate to time according to Kelsen’s theory. All these elements will help us to better
understand how central the temporal aspect is and how deeply it affects the legal validity in
Kelsen’s theory.
Before examining how the derogation works in cases of norms which might lose their validity
in time or because of a situation of conflict, let us first analyze how normative validity relates
to time in general.
97
7.1 Norms versus Sentences
It is never enough remaking that Kelsen’s principal methodological interest is differentiating
norms from sentences, as well as Law from Legal Science, and the problem he has in mind
when doing so is the common practice (exemplified by the works of, among others, Walter
Dubislav, Sigwart, Grue-Sörensen, Manfred Moritz or Ulrich Klug) of the application of logical
principles to norms, mainly the inference rule in the context of normative creation, and the
non-contradiction principle in the context of normative conflicts. Let us try to understand how
the validity of norms can come to an ending or simply expire with time (something not to
happen with the truthiness or falseness concerning sentences), and how Law dispose of
special means to deal with this temporal aspect of normative validity.
The norm is expressed by an act which can be made explicit by various means, such as
gestures, signs or imperatives. Norms are often expressed by imperatives, and don’t have
truth value: they can only be said to be valid. But they can also be described in sentences, like
the ones which form the legal science, and those descriptions about norms can be true, if they
affirm the fact of the validity of a norm, or false when the affirm the validity of a non-valid
norm (or the contrary). In Kelsen’s terminology, norms are said to be the sense of an act of
will, meaning that they are valid when the behavior they display is the result of someone’s
(the Legislator’s or the Judge’s) willing. In the same way, statements or sentences about norms
are simply the sense of an act of thought, which can be evaluated as true or false as any other
sentence.
The important point to us is that validity of norms and truth of sentences do not have the
same temporal characteristics.
First of all, the main difference between the two is that norms are equivalent to validity, while
the sentence displays the property of being true or false. The validity does not operate by
giving a classification or qualification of a norm, in the same way that we can classify or qualify
statements according to truth values, or at least that we can distinguish them according to
their truth values. The norm is valid in the moment of its creation or, still, the norm is created
already as a valid norm.
Then, in its creation, the norm is said to be valid for a determined space of time and territory,
and this depends on the creator’s willing. Differently, a true evaluated sentence, to Kelsen,
possesses in itself the property of being true, independently of our knowledge about it. So,
98
even if in the Middle Ages it was true to everybody that the Earth was the center of the
universe, this was always and will always be a false sentence about it. It was our access to the
real value of the sentence that was wrong. Similarly we can say that the statement affirming
the heliocentric theory was not false (or incorrect) during the period of time regarding the
middle Ages, but was and will always be a true or correct (except if one day in the future we
discover that this information turns out to be wrong). Actually Kelsen is saying that a
statement about a fact never alters its “original” truth value: it’s the agent who either knows
the truth value in question, either ignores it or makes a mistake concerning it.
Similarly, a false sentence is still considered to be a sentence, with the property or quality of
expressing a false information. It does not cease to be a sentence in the moment it is
discovered to be transmitting a false information. But the norm begins its existence from the
moment on it obtains its validity, and ceases to exist in the moment when it loses it. That’s
why there can never be an invalid norm, because it would not be an existing norm. Valid norms
exists – they are present in a legal order, their validity is fixed in time and space, they are
positive – and invalid norms do not exist, they could not come into existence in the first place
without the validity aspect.
Another interesting aspect concerning our problem consists in the fact that norms are not
always expressed by conditionals or hypothetical ought-propositions, but rather by
predictions, previsions or forecasts of a future event. So, when the norm is formulated saying
that the guilty “will be” punished by such and such means, this does not mean the prediction
of the application of the sanction, but it expresses, in a figure of speech, an imperative
prescribing the application of the sanction. Even under those formulation, the norm does not
predict the future, it has no informative sense, but prescriptive sense, and still can only be
said to be valid.
Another relation between the validity and the temporal aspect is given via the notion of
efficacy. So, the norm is valid once belonging to a legal order (via an act of will coming from
an authorized person), but another condition is that this legal order must be, in its overall,
also efficient.
So we can say that the efficacy of the order is a necessary (though not sufficient) condition for
the validity of the norm. Hence, the efficacy is the direct link between the norm and the
time/space aspect: by its efficacy we attest that the norm is observed or applied in reality, in
the temporal, territorial, material and personal spheres, including its retroactive effect. This
99
last element leads to an important aspect of the temporal features of normative validity: the
derogation.
7.2 Derogation
Besides commanding, permitting and authorizing, derogating is a specific function of the
norm, with the specificity that it only takes place in relation to another norm’s validity, which
will be terminated by the derogating norm.
First of all, we must notice that it is not only via the “active” function of derogating, but also
by the lacking of efficacy after a determined amount of time in which it has not being regularly
applied that the norm loses its validity, that is, ceases to exist. In this case, the loss of the
validity is explained by the lack of application of the norm during the time, and refers to the
human behavior directly. Also, a norm can lose its validity by derogation simply when the
authority understands that a determined norm is no longer desirable in the order. But the
most common context of the derogation function is in cases of conflicting norms. We will
come back to this special feature later106.
7.2.1 The Formulation of the Derogation Norm
The derogating norm is not expressed by an imperative, simply because it does not prescribe
a determined behavior or its abstention. The derogating norm is always directed to another
norm, which validity is about to be terminated. It assumes the form of an assertion of the kind:
“The norm according to which X ought to Y is, by this means, repealed.” Even if this assertion
doesn’t have an expressly prescriptive sense, it conserves its normative sense, for it is
abolishing another norm’s validity. Kelsen gives an example:
“If there is a valid norm ‘Males who are 21 years old and found fit for service
are to do military service’, and if the legislator were to posit the norm ‘Males
who are 21 years old and found fit for service are not to do military service’,
106 Section 7.3 in this Chapter.
100
he would not repeal the validity of the first norm but posit a norm which
conflicts with it. But a derogating norm does not conflict with the norm whose
validity it repeals. But it is not possible linguistically to formulate the
derogating norm in the following way: ‘Males who are 21 years old and found
fit for service not-are to do military service’. Hence derogating norms appear
in the form of statements, for example, ‘The validity of the norm ‘Males etc.’
is hereby repealed’. This statement does not have a descriptive but a
prescriptive function, i. e., that of repealing the validity of a norm.”107
Roman jurisprudence distinguishes between abrogatio as complete abolition from derogatio
as partial annulation of a norm’s validity. The complete abolition of a norm’s validity refers to
its substitution concerning for example another norm which the content will be more
restricted concerning the territorial, temporal or personal aspect. As to the partial annulation,
there is simply a change is the formulation of the normative content, which is not substituted
by another norm, but reformulated with a slightly different content. We must note that Kelsen
defends that, in the situation of normative conflict, one of the norms must be completely
abolished; otherwise we would still be in presence of a conflicting situation. That’s why, for
Kelsen, the derogation is a specific normative function abolishing completely another norm’s
validity. Kelsen explains that:
“The distinction between abrogare and derogare goes back to the celebrated
passage in Cicero’s Republic: ‘It is a sin to try to alter (abrogare) this law, nor
is it allowable to attempt to repeal (derogare) any part of it, and it is
impossible to abolish (abrogare) it entirely’ (iii.22). This passage clearly
concerns a statute made up of many legal norms. But the principle Lex
posterior derogat priori applies not only to the relations between statutes,
but also to the relation between single legal norms, and in this principle the
verb derogare means not only ‘partial’ repeal, but also ‘total’ repeal. I use the
word ‘derogation’ in the sense of the repeal of the validity of a norm.”108
107 Kelsen, H. (1979, p. 108). 108 Kelsen, H. (1979, p. 114).
101
7.3 The Context of Normative Conflict
Normally, the derogation takes place when two norms are in conflict and then, either the
validity of one norm is abolished and no new norm is created in its place, or a new norm is
then created.
A situation of conflict between two norms means that obeying or applying one of the two
implies, possibly or necessarily, infringing the other. But for the two norms to be in conflict,
both of them need to be valid. And here we can see again the difference between norms and
sentences: in the case of the norms, the derogation acts over one norm’s validity which, from
that moment on, ceases to exist; but in the case of the two contradictory sentences, one of
the two is expressing something false from the beginning. Concerning the norms, even the
descriptions of their validity are both true sentences, since the two norms are, by the time of
the existence of the conflict, effectively valid.
The normative conflict can, but doesn’t need to be solved by derogation, which is something
to be stipulated by a legislative authority. The derogation is not a logical or legal principle, but
a normative function, which can be executed also without the presence of conflict. But in the
conflict situation, it is the case of a third norm directed to one of the conflicting norm’s validity,
in order to abolish it. This conflict can take place between norms of the same level, or between
a norm superior and its inferior which were stipulated in different periods of time (one
anterior and other posterior). Besides, the conflicting norms must belong to the one and the
same normative order. So, if a moral norm is in conflict with a legal norm, the derogating norm
coming from Law can only derogate the legal norm, and not the moral norm, and vice-versa.
7.3.1 Some Examples
According to Kelsen, the derogation as a normative function belongs not only to the legal
domain, but also to positive Morals. Kelsen says:
“It can also occur in a positive morality: for example, in Christian morality,
when Jesus says: ‘Ye have heard that it hath been said, Thou shalt love thy
102
neighbor, and hate thine enemy. But I say unto you, Love your enemies…’
(Matt. 5: 43-4). That means: I repeal the validity of the norm ‘Hate your
enemies’ and I posit the norm ‘Love your enemies’.”109
But Kelsen also suggest that the most common in the Moral field is that a norm loses its validity
not by an “active” procedure of derogation, but rather by the lacking of efficacy of this norm
during time. On the legal field, one example that can be mentioned is the “Prohibition”
concerning the manufacture and sale of alcohol during the 20th in the United States, and which
has completely lost its efficacy at the point be formally repealed in 1933 by an official act: the
21st Amendment repealed the 18th Amendment to the US Constitution, by the method of the
state ratifying convention.
Another interesting aspect concerns the Roman law principle called lex superior derogat
inferiori, which is denied by Kelsen in its theory, since it is mistaken to think of one of the
norms in conflict as derogating the validity of the other. The same thing happens concerning
the context of conflicting norms from the same normative level, where Kelsen equally attacks
the Roman law formula lex porsteriori derogat priori. Kelsen says that “It does not also work
as a principle the formula lex posteriori derogat priori, since that it’s not correct that the
derogation comes from one of the conflicting norms. The derogation takes place only via the
interpretation.”110
Then, the reason why the formula lex porterior derogat priori as well as the formula lex
superior derogat inferiori are mistaken is because they give the idea that the derogation is a
function of one of the two norms in conflict. But those two norms are addressed to human
behavior, not to normative validity. On the contrary, the derogating norm only exists in
function of the norm to have its validity derogated: it comes into existence in the moment of
conflict, in order to solve it, and ceases to exist once the conflict is solved and its function is
performed.
Another reason why the formula of Roman jurisprudence lex porterior derogat priori is
mistaken it’s because it can be the first norm to be derogated, and also even both the
conflicting norms. The derogating norm is presupposed by the authority in question, who
109 Kelsen, H. (1979, p. 106) 110 Ibidem, p. 236.
103
assumes this norm in the moment of need, but it does not exist in the legal order among the
other positive norms.
7.4 Retroactivity
When a norm is posited as valid, its validity concerns a determined territory and period of
time, called the norm’s sphere of validity (which by the way can also be unlimited). Concerning
the temporal sphere, we must consider both the period of time before the norm is posited
and after it. This could seem strange in the sense that, normally, a norm regards the regulation
of the human behavior to take place after its existence, after the fact of being posited within
a determined legal system. Nevertheless, it is still an open question if a norm can be also be
directed to a behavior taking place before the existence of such a norm, and then we are
talking about the norm’s retroactive effect.
Most jurists will be against that legitimate existence of the retroactive effect of norms, to be
accepted only, and still, maybe, if there is benefit for the defendant or when the law explicit
indicates the effect.
These aspects are especially problematic when regarding cases of ex post facto laws regarding
war crimes, for example, but also in the case of, for example, a statute of a regime becoming
valid by a revolution which repeals the validity of the previous statute belonging to the
substituted government. By the new statute, certain behaviors previously considered to be
political crimes to be punished, will now be reconsidered. Kelsen says in a very clear manner
that “Admittedly, what is done cannot be undone, but the normative interpretation of events
long past can be changed retroactively by means of norms posited after the events to be
interpreted.”111
This is a complex matter and our intuitions indicate that Kelsen is mistaken in his approach to
retroactivity. In the case of individual norms, they are indeed directed to a behavior which
was performed before the enacting of the legal decision, but this does not indicate a
retroactive effect, simply because this decision represents the application of a general norm
already valid when the delictuous behavior took place. The legal decision only have retroactive
111 Kelsen, H. (1979, p. 145).
104
effect when no general norm is applied, because there was no valid general norm concerning
the case when in has occurred, i.e. if the behavior was not considered a delict by Law by the
time it has occurred. But, as we have said, the retroactive effect is normally fully denied in
most Countries, and Kelsen’s inclination to it doesn’t seem to be legitimate or defendable.
7.5 The Basic Norm
Now that we’ve seen how a positive norm’s validity relates to the temporal aspect, let’s see
how this issue affects a normative fiction by excellence, which is in Kelsen’s theory
represented by the notion of Basic Norm.
According to Kelsen, there are norms which are not the meaning of an objective act of will,
but merely the meaning of acts of thought. For example, I can think of a norm such as “All
people are not to eat meat”, but without wanting such a norm. The point is, nothing defends
me though to think of this norm, even if it does not exists, if it is not a positive, binding norm
on every people’s behavior.
Kelsen’s point is that, similarly to this case, there can be also norms that are the meaning of a
fictitious act of will – they have a fictitious validity: we think of those norms as if they existed,
as if they were valid, and one normally does that if there’s some purpose or interest regarding
the “existence” of such a norm.
As we have already seen, the notion of Basic Norm as a fiction is inserted in the discussion
about the grounding of the legal positive order. In order to create a norm we need, first, of a
more general norm with a correspondent content and, second, of an authority figure whose
act of will will have the norm as its meaning. Given this context, one rapidly recognizes that
the fact of going up in the normative hierarchy for the searching of the normative justification
will lead us to the top of the normative pyramid illustrating the legal system. Then, what we
will find on the top of it will have to be the Constitution, or, if we want to go even “higher”, it
will be the First Constitution of a Country. But, at that level, we are pushed to argue about
what is giving the legitimation of the content of such a document and, even more, who could
have given to the Legislator the authorization to elaborate such a binding document like the
Constitution of a Country…
105
At this point Kelsen suggest that the only solution to this problem would be find if we
presuppose a fictitious norm, which is not posited since this positing would again postpone
the problem of the asking for the justification of such an existing norm as well as its author.
Rather than this, we must simply make as if there were a norm higher than the Constitution,
giving its legitimation. This presupposed higher norm would be the fictitious Basic Norm of
each legal system.
Now, analyzing the relation of the Basic Norm’s validity with the temporal aspect is a difficult
issue, since the Basic Norm is not a valid norm. To be valid means to exist, to be posited, to
be a part of the legal order. Rather, the Basic Norm is a thought norm, and we make as if it
was valid. The Basic Norm is not even formulated, since to give content to it would mean
having to justify it by a higher norm, which is exactly what we are trying to avoid. The great
abstractive feature of this fake-norm detaches it from the spacious-temporal sphere.
In the present study we defend that, even if the Basic Norm cannot be recognized as a part of
the legal system, it is not by so present in the legal system. We see the Basic Norm rather as
a scientific tool, used by the legal scientist when is question to justify the legal order.
But concerning the temporal aspect of legal validity, it is still acceptable to think that, for
example, every Constitution must be considered along with the idea of a fictive Basic Norm
which justifies it (along with the whole legal system in consideration), and even that, if a
Constitution is derogated, let’s say because of a revolution, a new fiction of Basic Norm follows
the new Constitution taking place. This perspective allows us to defend that the fictive validity
of the Basic Norms will be as limited in time as the valid Constitution to which it is directed to
scientifically justify.
7.6 Final Remarks on this Topic
With this exposition, our main objective was to situate the normative validity in relation to
the temporal sphere. First, we analyzed the importance of differentiating valid norms from
true/false sentences, and how both validity and truth values relate to time in different
manners. Then we saw how the derogation is not a legal principle along with some Roman
law formulas, but rather a normative function, having as main context situations of normative
conflict. We also saw that the norms can lose their validity not via the derogation, but simply
106
because of their lacking of efficacy over time. Afterwards, we studied how the normative
retroactive effect, when admitted, affects the validity of norms and, at last, we saw that
normative fictions do not relate to the temporal sphere in the same way as norms, since they
do not have positive validity.
All these aspects show that the temporal sphere of validity plays a central role in Kelsen’s legal
theory. It helps us to better understand the particularities of the legal domain, and most of all
it helps justifying Kelsen’s worries in preserving the limits of Law from external elements, such
as logical principles. In this sense, the specific relations between legal norms and the temporal
aspect can then be seen as another evidence or another argument in favor of Kelsen’s
methodological “purity” of the legal theory.
Perhaps the main collaboration of this chapter was to show and explain that, since norms do
not relate to their validity in the same way as sentences relate to their truth values, the
situation of conflict cannot be compared to the situation of contradiction. This also represents
the complete refusal of the possibility of application of the principle of non-contradiction to a
situation of conflict between norms.
107
Chapter 8
The Notion of Practical Reasoning (Part I)
“Under normal conditions the research scientist is not an innovator but a solver of puzzles,
and the puzzles upon which he concentrates are just those which he believes can be both
stated and solved within the existing scientific tradition.”
Thomas S. Kuhn
In this chapter 8 we propose to introduce the first part of a logical frame allowing us to display
the argumentative features behind legal decisions and the justification provided by legal
science. This undertaking is motivated by Hans Kelsen’s solution to a well-known puzzle in
legal philosophy, called Jørgensen’s dilemma, based in the neutral element of the modally
indifferent substrate.
In this Part I, we deliver a detailed presentation of the problem, as well as two of the many
attempts to a solution. Then we present what we consider to be the final solution, given by
the legal philosopher Hans Kelsen.
Based on this approach, the next chapter (Part II) will serve to introduce our attempts to
provide, by means of a dialogical frame, an original application of the Kelsenian solution in the
field of legal justification. This logical frame we propose not only perfectly displays Kelsen’s
approach but it also allows to express, debate and justify the legal reasoning without
transgressing the limits between the legal field of normative creation and the scientific field
of normative justification.
108
8.1 Introduction
The closeness between the domains of law and philosophy is corroborated by the prolific
discussions concerning the foundations of the theory of law, namely the questions involving
the definition of a legal norm. In a wide sense, norms are orders, commands deriving from
human will; but legal norms specifically are obligatory imperatives, binding prescriptions, they
are an “ought” (Sollen). According to Kelsen, there is a “methodological abyss”, a gap between
the “Ought” (Sollen) domain, where we find the norms, and the “Is” (Sein) domain, where we
find the sentences, which are the result of an act of thought. Jorgensen’s dilemma puts in
question which kinds of interaction seem to prevail in the relation between prescriptions
(Sollsätze) and descriptions (Sätze)112.
This dilemma, as pointed out by Giovanni Sartor, “consists in the supposed necessity of making
the following choice: Either one rejects using logic in the law (and more generally, in practical
reasoning) or one has to admit that legal contents (practical noemata) can be true or false”113.
In other words, the dilemma deals with the question whether imperatives can be a part of a
logical inference, as one of the premises or as the conclusion. If a first reaction points to a
negative answer (since norms cannot be true or false114), most of us will have to agree that
inferences involving norms make part of our daily lives and that the evidence of their legal
validity115 seems to be out of question. This difficulty in understanding the role of imperatives
in our reasoning about norms intrigued also Hans Kelsen, who severely criticized the answers
previously given.
First of all, our aim is to explain how Kelsen deals with the dilemma, by arguing that there
cannot be any logical derivation between imperatives (norms) and sentences. Then, we will
show how Kelsen’s approach offers a theoretical frame to a brand new treatment of problems
like Jørgensen’s dilemma and consequently helps us to provide an original answer to our first
problem, namely: How can deontic logic be reconstructed in accord with the philosophical
position that norms are neither true nor false?
112 This distinction will also be important in the differentiation between legal norms (Rechts-Norm) and normative propositions (Rechts-Satz). 113 Cf. Sartor, G. (2005, p. 420). 114 Norms are not facts, they are the “product” or the sense of an act of will, they are wanted by someone, and directed towards someone else’s behavior. 115 Here legal validity means the specific existence of a norm in a legal order. For a norm to be valid or existent in the legal order means that all the legal procedures to its creation were respected.
109
In order to do so, we reconstruct the relations between law and legal science using a particular
approach of logic, namely the dialogical approach116. Not only this framework allows us to
respect the methodological abyss pointed out by Kelsen, but it also displays the interactions
which are inherent to the legal reasoning. Our attempts are particularly directed to aspects
concerning the justification of legal decisions, which will be in the following chapter
considered within a logical frame, as a means to try to put in evidence the argumentative
process involved in the legal reasoning.
8.2 Understanding Jørgensen’s Dilemma
Jørgensen’s dilemma is inserted in the discussions concerning normative reasoning and
deontic logic. It states the problematic fact that, even if it is commonly accepted that norms
cannot receive logical treatment since they are not true or false, it seems to be legitimate to
derive a specific norm from a general one, through a rational operation such as a syllogism.
This syllogism should then be called “practical”, because it contains norms as the major
premise and as the conclusion.
The existence of practical syllogisms would affect many aspects of our understanding, dealing
with questions going from the rationality of the normative discourse to the justification of
moral acts and decisions, crossing controversial aspects such as the conflicts between norms,
moral dilemmas and the possibility of a logic of norms.
8.3 First Attempts: Jørgensen and Ross
The easiest way to understand this problem, first announced by Jørgen Jørgensen in 1937, and
named after this author by Alf Ross some years later (1944), is to immediately invoke some
examples.
116 Presented in the next chapter.
110
In our daily lives, we are frequently confronted with legal norms and imperatives in general,
and we are demanded to ratiocinate and theorize about them. Most frequently, we do so
without paying attention. Jørgensen’s classical example is:
Keep your promises.
This is a promise of yours.
Keep this promise.
The question imposed considers whether imperatives can be a part of a logical inference, as
a premise or as the conclusion. Since norms cannot be true or false, it would seem reasonable
to answer negatively. However, most of us will agree that practical syllogisms of the form of
the example above seem to be legitimate, even if we have no criteria to distinguish valid
practical syllogisms from invalid or arbitrary ones. This difficulty in understanding the role of
imperatives in our reasoning about norms, commands, orders and in the legal argumentation
was called by Alf Ross the “Jørgensen’s Dilemma”.
Let us analyse some examples given by Jørgen Jørgensen and Alf Ross, indicating the fact that
the notion of practical syllogism is misleading and erroneous. Following a Kelsenian approach,
the primal obstacle to the logical treatment of imperatives in a practical syllogism lies on the
fact that they are not the objective sense of an act of will.
8.3.1 Jørgensen’s Answer to the Dilemma
The main problem concerning Jørgensen’s dilemma does not concern the fact of deriving
norms from facts (or indicative sentences), but consists in logically derive a norm from another
norm117. His first reaction to this problem is to say that imperatives (norms) cannot be a part
of a syllogism, because they cannot be true or false. But then he proposes another approach
to the logical treatment of those imperatives: Jorgensen says that the answer to the
conundrum lays in the difference between the imperative factor and the indicative factor
present in the norm.
117 Cf. Cabrera, C. A. (1999, p. 207).
111
Since there is no imperative without “something” which is demanded to be performed, there
is always a possibility to detach the object of demand from the imperative. In this way, the
imperative factor concerns the expression of the subject’s state of mind: the act of
commanding, the act of giving an order. The indicative factor concerns the specific content of
the norm and has a propositional feature, and thus it may be treated with some logical
tools118.
In order to be able to logically manipulate the imperative we must, first of all, subtract the
indicative factor from the imperative. So, from
“Tax evasion must be punished by fine and/or imprisonment”
we simply subtract the indicative sentence
“The norm demands that ‘Tax evasion is punished by fine and/or imprisonment’”
and, in a simpler way, we can directly convert it to the indicative sentence
“Tax evasion is punished by fine and/or imprisonment”
which can then be evaluated as true/false. Now, since evidence shows that Jérôme C. actually
practiced tax evasion, we are allowed derive another indicative sentence, namely,
“Jérôme C. is punished by fine and/or imprisonment”.
From that, we go the same way back by adding the imperative factor and reformulating a
norm such as
“Jérôme C. must be punished by fine and/or imprisonment”.
To Jørgensen, the translation procedure explains the efficacy of the practical syllogism, in the
same way as it justifies the inference process. The syllogism is, according to him, legitimate,
since, in fact, the inference is made concerning the indicative factor, and the norm is obtained
by the reconstruction of the imperative factor.
8.3.2 Ross’ Answer to the Dilemma
118 This form of distinction can be found later with the neustic/phrastic dichotomy in Hare (1972).
112
Ross severely criticizes Jørgensen’s answer. The problem, says Ross, is that we are not able to
reconstruct the norm “Jérôme C. must be punished by fine and/or imprisonment” (I2 in the
figure below) from the first norm “Tax evasion must be punished by fine and/or
imprisonment” (I1 in the figure below). Jørgensen’s logic of satisfaction is useless to explain
how we derive new norms as conclusions in the practical syllogisms119. In the figure below,
Ross illustrates the gap through this illustration, where “I” stands for “Imperative” and “S”
stands for “Sentence”:
I1
S1
?
I2
S2
So, concerning the dilemma, even if Jørgensen says that we can logically approach the
indicative factors of the norms involved in the practical syllogism, still there is nothing allowing
him to say that we can derive a new imperative from the first one stated in the major premise,
even if this procedure is made in an indirect manner. Therefore, the dilemma remains
unsolved.
To solve Jørgensen’s problem, Ross suggests several approaches. First, he suggest an
alternative for the apofantic thesis that logic only deals with true/false valued sentences, by
saying that the values true/false can be substituted by validity/invalidity, which now should
be seen as the logical value of the norms but in a subjective way120.
119 Dependent on the efficacy of the norm. Ross explains: “[. . . ] an imperative I1 is said to be satisfied when the corresponding indicative sentence S1 , describing the theme of demand, is true, and non- satisfied, when that sentence is false.” See Ross, A. (1944, p. 37). 120 We will analyze this misconception concerning the validity as a value of the imperative or norm later on this same section 8.3.2.
113
Ross understands validity as defined by “the presence of a state of mind in the person, that
determines this validity”, and depends on the will of the imperant. The validity turns into a
psychological, not a semantic concept.
Ross starts his paper by stating the problem as being:
“[. . . ] to elucidate whether sentences which are not descriptive, but which
express a demand, a wish, or the like, may be made objects of logical
treatment in the same or a similar manner as the indicative sentences.”121
Then, Ross proposes a wider delimitation of the logical domain, by suggesting the substitution
of the logical values of truth and falsity by validity or invalidity. So, to fill up Jorgensen’s gap
in deriving I2 from I1 (See the next figure below), Ross declares the following:
“It may then be laid down: if there be any sense in ascribing objective validity
and invalidity to imperatives or to a certain group of imperatives, then it is
possible to interpret the logical deductive system as being applicable to those
imperatives. The logical deduction of I2 from I1 then means that I2 has
objective validity in case I1 has objective validity.”122
But how can we verify the validity/invalidity? Ross answers that the process of objective
legitimation ignores the person and focuses only on the impersonal norm, and this will finally
lead to religious morals or to natural law kind of doctrines (imperatives without imperator).
I1
S1
I2
S2
121 See Ross, A. (1944, p. 31). 122 Ibid, p. 35.
114
Then Ross considers a second possibility, saying that the means of verification of the validity
lies on its satisfaction, so that we would have a perfect parallel between an imperative and its
satisfaction. In this way, from the imperative “Close the door” we may have the correspondent
sentence “The door is closed”. Continuing the parallel, we must be capable of having also, for
example, the negation of the sentence, i.e. “The door is not closed”, from which we should
attain the correspondent imperative “Don’t close the door”. So, in the inference, when we
have “Close the door” satisfied, we must have also “The door is closed”. Therefore, since the
negation is false, its correspondent “Don’t close the door” is not satisfied.
This example shows the complete lack of similarity between the prescriptive field and the
logical field. Moreover, similarly to the negation, a disjunction could be added to the
imperative, as in the famous example of “Slip the letter into the letter-box” to which we add
the disjunction as: “Slip the letter into the letter-box or burn it”. Even if the adding of the
disjunction shows no problem concerning the truth values of the correspondent sentences, it
is not evident, and even not plausible, that such a disjunction might be relevant in the
normative domain, concerning the imperative form (the obligation of sending the letter
doesn’t imply the obligation of sending it and burning it—according to the inclusive
disjunction—because the obligation of sending it is not satisfied if we burn the letter). These
examples show precisely how deontic logic cannot grasp the relations between imperatives,
and the analogy with the satisfaction is clearly not a good alternative.
Ross then finally proposes the third solution to the problem. The logical aspect of the
imperatives in the practical inference concerns not the satisfaction or the verification, but the
“subjective validity” of those imperatives. In this case, says Ross: “An imperative I1 is said to
be valid when a certain, further defined psychological state is present in a certain person, and
to be non-valid when no such state is present.”123
The condition here is that the double validity/invalidity is defined by the truth/falsity of the
corresponding indicative sentences, that is to say, the logic directly applied in the true/false
valued sentences would be indirectly applied in the valid/non-valid imperatives when in
presence of a determined psychological state. This new element would prevent us from the
problems concerning the previous examples such as “Slip the letter in the letter-box or burn
it!”. The “psychological state” which Ross invokes would serve as a justification or a motivation
123 Ibid., p. 38
115
to the imperant’s will, avoiding trivialities. But this approach is doomed to refer always to a
specific case (or a specific trial, for instance) where a norm or an imperative were applied
regarding a specific case. Again, the normative relations are put aside, because it is the fact of
the creation of the norm that is being asserted in a statement. As we will see next, according
to Kelsen’s approach, these kind of statements actually compose the science of law, i.e. they
are about the validity of the norm, for instance “There was enacted a valid norm saying that
‘Murder is punished by imprisonment”’, or “It is true that the norm ‘X’ was applied by Judge
Z in case Y”. Again, this can be logically treated (in the same way that legal science can be
logically treated as any other science), but this does not take into account any relations
between norms.
So, according to the transformations between imperative and indicative factors allowed by
the rules of satisfaction, the validity of the first norm would be then preserved in the
conclusion. But, since nothing guarantees that the person enacting the first order must also
enact or want the specific order in the conclusion (in other words, there is no logical—nor
legal—necessity between the two norms)124, a psychological element is required. With this,
Ross concludes that:
“Imperatives can be constituent parts of genuine logical inferences, but if so,
it is simply a question of a “translation” of logical inferences concerning
indicative sentences about the psychological facts which define the “validity”
of an imperative.”125
So, Ross concludes that Jørgensen’s dilemma is the result of a pseudo-logic puzzle. Even if
Ross’ approach is not restricted to the legal domain, but rather to the prescriptive domain in
general, the problem of the legitimacy of the created norm remains the same. Besides, the
author recognizes that he does not “solve” the puzzle, but rather shows that it is a pseudo-
problem or, more precisely, the case of the use of some kind of pseudo-logic. He says that we
simply consider as being a true evaluated statement the fact that a norm has been applied by
a judge, and obeyed by a person, and the valid imperatives would be transformed in verifiable
sentences such as “The judge X has enacted the norm ‘Y”’, or “This person has accepted the
124 The point here is that the specific norm in the conclusion depends on the act of will of the person enacting such a norm, and this process is not achieved through a syllogism. The judge can enact the particular norm without making use of the practical syllogism. 125 Ibid., p. 45.
116
norm ‘X”’, which does not concern at all the prescriptive level anymore. It no longer represents
a practical syllogism, because the norms are being simply described as having or having not
taken place. That is why Ross calls it the application of pseudo-logic.
This discussion deals with many other situations of norm-creation, when the norm is enacted
not by the judge or the legislator, but also by any other “authority”, such as a father facing
his son, or a teacher facing his/her students, or even cases of decision-making which could be
reconstructed via a practical syllogism.
In the end, all these cases reflect a complete independence of the norm pretending to be the
conclusion of the syllogism with respect to the norm admitted as a first premise. This
correspondence has no effect on the preservation of the validity from the general norm in the
premise to the specific norm in the conclusion. Or, in another words, the imperative in the
conclusion is not legitimated or binding because of the departing imperative. And Ross himself
recognizes that that is where the dilemma emerges: as the result of a psychological
mistreatment of the norms.
8.4 Kelsen’s Battle against the Dilemma
The works of Hans Kelsen are marked by the constant and close relation to philosophical
questions, such as those involving normative justification. It is even a common saying in the
juridical area that Kelsen was the greatest philosopher among the jurists and the greatest
jurist among the philosophers of his time. So, the fact that he devoted a great deal of his
attention to the problems concerning practical reasoning and more specifically to the
Jørgensen’s dilemma is not surprising.
In the next sections we will analyze how Kelsen treated the previous answers to this problem,
and what was his particular solution to it. We must note that Kelsen’s approach is marked by
his principle concerning the existence of a methodological abyss between the domains of the
“Ought” (the normative domain, formed by valid norms, accessed by acts of will—the domain
of law) and the “Is” (the descriptive domain, formed by true/false sentences and descriptions,
accessed by acts of thought—the domain of science). This conception is essential to
understand how, similarly to Ross, Kelsen manages to show that Jørgensen’s paradox is an
illusion.
117
8.4.1 Correcting Jørgensen
To begin with Kelsen’s treatment of Jørgensen’s Dilemma, we must at first understand what
this author means by the term “modally indifferent substrate”.
When considering the differences between the norms (“ought” sentences) and descriptions
of norms (“is” sentences), Kelsen insists on the fact that the two cannot be reducible or even
comparable. One is the law, and the other is the science that describes the law as its object.
Kelsen says:
“‘Is’ and ‘Ought’ are purely formal concepts, two forms or modes which can
assume any content whatsoever, but which must assume some content in
order to be significant. It is something which is, and it is something which
ought to be. But no specific content follows from the form.”126
So, if we consider the following sentences:
1. “Jérôme C. ought to pay the taxes.” and
2. “Jérôme C. pays/had paid his taxes.”
We have one and only one modally indifferent substrate, which is “To pay the taxes”, under
two different modes, namely, the “ought” in 1. and the “is” in 2. Their content is identical, but
the sentences themselves cannot be comparable, correspondent or deducible one from
another.
So, concerning Jørgensen’s solution to the dilemma, if we stick to Kelsen’s conception of the
modally indifferent substrate, we must notice that, in fact, the modally indifferent substrate
is neither true nor false in itself. It can be the content of a sentence to be evaluated as true or
false, in the same way that it can only be the content of a valid norm. But there is no
correspondence between a true/false sentence and a valid norm; they only share the same
modally indifferent substrate. So, when Jørgensen says that the answer to the dilemma is only
a matter of translation, actually there is no correspondence between an imperative and an
126 See Kelsen (1979, p. 58).
118
indicative sentence: the fact is that we are dealing with the same modally indifferent
substrate, presented under two different modes. To sum up, the point of interest is that the
modally indifferent substrate is outside the bounds of any logical approach.
8.4.2 Correcting Ross
As we have seen, Ross’s approach takes as a point of departure the fact that the validity is the
value of the norm (or of the imperative) in the same way that truth/falsity are the possible
values of the indicative statement or a sentence. So, even if Ross’s conclusion is that the
dilemma is not legitimate since some pseudo-logic is being applied to the imperatives, all his
argumentation lies in the presupposition that, when considering practical inferences, the
validity is a value of the imperative in the same way that the truth is a value of the indicative
statement. A nowadays critic will resume that: “Ross directly attacks the idea that the validity,
a specific quality of the prescriptive propositions, would be equivalent to the truth, a quality of
the indicative propositions.”127
So, even if Ross’ solution is closer to Kelsen extremism in denying the possibility of logical
treatment of the norms, Ross takes the wrong way to arrive at this same conclusion128.
Indeed, Kelsen will perfectly agree that they are not equivalent at all, but, more than that,
Kelsen would also say that the validity is not even a value of the norm. Moreover, still about
the differences between norms and statements, we must note that, when considering this
cornerstone of Kelsen’s legal theory, we immediately recognize that, actually, the validity can
never be predicated from a norm in the same way that the truth is a property of a sentence
to be evaluated as so. As we have already mentioned, the validity cannot be considered to be
a property of the legal norm, because it is the very fact of the norm’s pertinence to a legal
system. The norm can only exist, it can only be considered as long as a valid norm. Moreover,
127 The original, in French: “Ross attaque directement l’idée selon laquelle la validité, qualité spécifique des propositions prescriptives, serait équivalente à la vérité, qualité des propositions indicatives.” See Champeil- Desplats, V. (2002, p. 34). 128 As we remember, Ross says that the logical treatment in a practical inference is an illusion, since the imperatives are actually treated as indicative statements (because the evaluation concerns the satisfaction of the norm, which is a verifiable fact).
119
while considering positive systems, this is the case for any normative system, including
positive morals, and not only positive law.
Besides, as we ave also already explained, the main difference between a statement (Satz)
and a norm (Sollsatz) is that the first is the result of an act of thought, and can be evaluated
as true or false, and the second is the sense or the result of an act of will, and can only be
valid. This main distinction allows Kelsen to give a more precise treatment than Ross with
respect to the impossibility of deriving a binding imperative from a previous one: the “second”
is the result of nothing but an act of will, whose sense is completely independent from the act
of will giving place to the first norm or imperative. This leads Kelsen to explain that, even if
the two imperatives display the same content (the same modally indifferent substrate), the
acts of will must be two, neatly separated and independent from each other.
Before moving to the next section, it is important to note that it is clear from Ross’ writings
that he does not pretend to approach the practical syllogism from a legal perspective, as
Kelsen does, but rather from the perspective of our daily situations of decision-making, or
when confronted to imperatives in general. The problem, in this case, would not be the issue
of the legal objective validity of the norm “obtained” in the conclusion of the practical
syllogism, but rather its legitimacy, i.e. the fact of its bindingness given the first premise. The
same question is posed in both approaches, namely, what renders this imperative in the
conclusion obligatory or binding in relation to the first norm presented as a premise? Why do
I, once having accepted the first imperative, must also obey the second?
8.4.3 Kelsen’s Final Solution
So, after having seen Kelsen’s position in front of the two answers given to the dilemma, how
would Kelsen himself solve this puzzle? The answer lies in the difference between an act of
thought and an act of will.
To Kelsen, every norm must have content, the “modally indifferent substrate”, which is not
the indicative factor as Jørgensen believed, but the behavior wanted by the imperator. So, in
one way, the person enacting the norm knows the content of the norm, his will is directed to
some behavior, indicated by the norm. On the other way, the addressee of the norm must
also understand what is he supposed to do, how is he supposed to behave, and he does that
by observing what is the content of the norm. Let us see how it works in an example. Consider
120
the following case, a more extended version of the “Al Capone example” we mentioned when
first introducing these differences.129
Every citizen should pay the taxes. This applies also to, let us say, the French citizen Jérôme C.
Once a year he receives a letter saying how much he has to pay to the government, as income
taxes. The legal norm behind the letter, put in a simpler way, says that “Every citizen must pay
his/her taxes”, even if this very prescriptive element is not explicitly formulated in the text
addressed to Jérôme C. The norm-positing subject can justify his demanding for the money by
enumerating the benefits to be seized by the citizens, like the improvement of social and
public services, the reduction of budget deficits, etc. In Kelsen’s theory, those are to be seen
as acts of thinking preceding the norm, and they have nothing to do with the will to have the
taxes paid. They have nothing to do with the modally indifferent substrate, which is the
behavior itself, in this case, “paying the taxes”.
But let us see how Jérôme C. could be confronted to the imperative directed at him, namely
“Every citizen must pay his/her taxes”. Once confronted to this general norm (the first premise
in the misleading “practical syllogism”), Jérôme C. can have acts of thought which allow him
to understand what is he supposed to do, given his specific situation. He can think about things
like “If a pay my taxes, my equals will benefit from the improvements in education, security,
health care and so on”. Also, he can consider that “If I don’t pay the taxes, and if I figure out
how to keep it secret, I can benefit myself from my own money. . . ”. Outside the motivation
frame, he must first of all understand that he is obliged to pay the taxes: the addressee of the
norm must know what he is supposed to do. So, he also has acts of thinking which allow him
to comprehend things like “it’s me and no other person who has to pay these taxes”, “I’m
supposed to pay what it’s said in the letter, not more nor less”, and those are true/false
evaluated statements, those are the sense of acts of thought. They don’t belong to the norm
in question; they are about the norm enacted. After understanding all that, Jérôme C. can
accept that, once “Every citizen must pay his/her taxes” is valid (the sense of someone’s act
of will), and after the arriving of the letter from the tax collection, he must pay his taxes, i.e.
he must perform himself the general norm which was, by the letters sent to him, applied to
his case, understanding that “Jérôme C. must pay his taxes”—otherwise he will be in trouble.
This concerns a fundamental division between thinking about possible norms, and wanting a
norm, i.e. enacting it.
129 Cf. chapter 5, section 5.2.
121
Concerning Jørgensen’s approach, Kelsen says that the indicative factor pointed by this author
has in fact a complete different meaning than the imperative factor. The indicative factor (the
statement) is the outcome of an act of thought; the imperative factor (the command) is the
sense of an act of will, and the two are completely independent from each other.
This is how Kelsen solves the dilemma. The misleading “practical syllogism” actually works in
our daily lives because we are in fact reasoning about the norms involved in the inference,
through acts of thought. It’s natural that we try to understand what the norm is demanding,
in order to know how to behave. The minor premise which poses a fact is actually the context
in which we reason, and the major premise (indicating a norm in the practical syllogism) poses
the norm that we must understand, comprehend, which presents the behavior wanted. If I
understand the first premise in the context of the second premise, I’m able to know which
behavior I’m supposed to have. All this procedure does not concern the normative level. Also,
it’s perfectly possible to understand the content of a general norm and not want to observe
the behavior posed as obligatory. The particular norm posed as a conclusion in the practical
syllogism will only be a legitimate valid norm after being also the meaning of an objective act
of will.
8.5 Any Objections? Von Wright’s Deontic Logic
When considering von Wright’s first attempts to construct a logic of/for norms, we are
confronted with definitions of the following type:
If an act is not permitted, it is called forbidden. For instance: Theft is not permitted, hence it
is forbidden. We are not allowed to steal, hence we must not steal130.
What von Wright tries to do is to use logical equivalence to establish the three deontic modes
(obligatory, permitted and forbidden). In fact, given a primitive modality P (standing for “it’s
permitted that. . . ”) and an arbitrary proposition ϕ, it is easy to define 1. a modality forbidden
as ¬Pϕ , i.e. it is not permitted that ϕ , so ϕ is forbidden, and 2. a modality obligatory as ¬P¬ϕ
, i.e. it is not permitted that not ϕ , so ϕ is obligatory. This is exactly what von Wright does in
his example: “forbidden” is equivalent to “not permitted”. But Kelsen says that the only
normative function is the mandatory function, norms have no permissive function. When von
130 See von Wright (1951, p. 3).
122
Wright says that “We ought to do that which we are not allowed not to do”, he defines the
obligatority of an act by the negation of the permission to not doing this act131. Norms start
from the opposite, when something is commanded (mandatory normative function), the fact
of abstention is qualified as not permitted. Everything we need is only the mandatory
function, everything else is accessory.
Let us see what Kelsen himself says about it:
“Wright says in this connection: ‘If the negation of an act is forbidden, the act
itself is called obligatory. For instance: it is forbidden to disobey the law, hence
it is obligatory to obey the law’ [. . .]. Since the normative function being
considered is that of commanding, and since to forbid is to command an
omission, things are being stood on their head when the being- commanded
of an act is presented as the being-forbidden of its omission. ‘It is commanded
to obey the law’: if we want to make use of the concept of forbidding, we can
express the same idea by saying ‘It is forbidden not to obey the law’. But things
are back to front if we say ‘It is forbidden not to obey the law, and hence it is
commanded to obey the law’.”132
Kelsen is right in saying that it seems strange to define what is obligatory by the negation of
what is not permitted. But from a strictly logical point of view, if what is permitted can be
defined by what is not obligatory, conversely, what is obligatory can be defined by what is not
permitted not to do. It seems to be a tension between logical and legal interests. Moreover,
in the introduction of von Wright’s book133 , which includes a re-edition of “Deontic Logic”,
the author says that, about this last essay, a further refinement is needed in the treatment of
this question:
“Another application is to the logical study of the norms (normative
discourse). This latter study is important to ethics and the philosophy of law.
But it must be pursued with much more refinement than in my first paper
(here republished) on deontic logic. Philosophically, I find this paper very
unsatisfactory. For one thing, because it treats of norms as a kind of
131 See von Wright, G. H. (1951). 132 See Kelsen, H. (1979, p. 322). 133 See von Wright, G. H. (1957)
123
proposition which may be true or false. This, I think, is a mistake. Deontic logic
gets part of its philosophic significance from the fact that norms and
valuations, though removed from the realm of truth, yet are subject to logical
law”134.
This confession shows that von Wright actually commits the same mistake as his predecessors,
Jørgensen and Ross. When he says that, even being “removed from the realm of truth”, norm
are still submitted to logic laws, he is actually stipulating some kind of strange correspondence
between truth and legal validity.
8.6 Final Remarks on this Topic
In this chapter, we have just saw how Kelsen deals with problems derived from the falling into
the naturalistic fallacy, such as Jørgensen’s dilemma. As we also have seen, the means for the
non-logical treatment of Jørgensen’s dilemma lay on the notion of modally indifferent
substrate, also highly dependent on the notions of act of thought and act of will, which we
saw in detail in also the previous chapters.
Now, since we already know that it’s not possible to treat the legal norms with principles of
bivalent logic such as the principle of non-contradiction, the rule of inference, or the notion
of practical syllogism, there’s nevertheless another logical means to deal with this process.
Whoever the candidate would be, it has to respect the methodological profile Kelsen draws
for his legal theory. Besides, it is never enough to mention that any legitimate logical
treatment of the legal field will be restricted to the scientific level, and not to a direct
application to legal, moral norms or imperatives.
The methodological profile abovementioned will preserve the legal field from comparisons,
translations or reductions to the indicative field, and rejects the insertion of truth values
regarding norms. And that is simply because it is now attested that every approach that didn’t
observe those scientific precepts ended up failing in their purposes. In the next chapter of
the present study, we will explain how such a prudent and cautious approach can render a
134 See von Wright, G. H. (1957, p. vii).
124
coherent and relevant explanation of the process of legal creation and the justification of
norms.
125
Chapter 9
The notion of Practical Reasoning (Part II)
“If we will disbelieve everything, because we cannot certainly know all things, we shall do much what as wisely as he who would not use his legs,
but sit still and perish, because he had no wings to fly.” (John Locke)
Kelsen’s approach to Jørgensen’s dilemma gives us some insights about how to deal with the
justification of legal decisions in the legal argumentation.
A legal decision involves a general norm that is applied by the judge in relation to a specific
case (fact) in order to “create” a specific norm for that case. We all know now that the creation
of a legal norm depends on an objective act of will coming from an authorized person, and no
logical element interferes in this procedure. If we are about to deal with the normative
reasoning (or the normative argumentation), the fact of trying to compare and correspond
(legal) norms to (logical) propositions will forcedly lead to error. But with the notion of modally
indifferent substrate, Kelsen shows that it is fairly possible to reason about norms, to evaluate
our behavior in relation to a norm, without disrespecting the limits between these two
domains.
This is precisely the purpose we pretend to develop in the present chapter135. We aim to show
how is it possible to logically discuss and reason about norms, and still maintain the dichotomy
between law and its science.
135 This chapter represents a cooperative work between the author of this thesis and Dr. Sébastien Magnier, who hereby receives also the expression of the author’s full gratitude. The joint work will be object of a forthcoming publication. Cf. Sievers, J. & Magnier, S. (2015).
126
9.1 Dialogs about Kelsen’s Solution
If we were to draw a scheme concerning the modally indifferent substrate (figure below), we
must insist on the fact that this notion allows a double treatment of the normative content
(the behavior “paying-taxes”, as in our example) in terms of a (possibly) valid norm in the same
way that it might assume a form of a sentence to be evaluated as true/false.
In the first case the dialogical approach to logic will deal with the normative construction in
terms of a Procedural justification, while in the second case the treatment will be made in
terms of a Factual justification136. The main point is that both the legal and the logical
treatment respect and maintain the distinction between the two forms which can assume the
same content.
Norms (Ought) Sentence (Is)
“X ought to pay his taxes.” “ Jérôme C. pays his taxes.”
Valid True or False
Procedural Justification Factual Justification
Regarding the process of normative creation, Jørgensen and Ross insist in considering the legal
reasoning by forcedly comparing, corresponding, translating sentences into imperatives and
vice-versa. In the end, this results in disturbing structures such as the practical syllogism. But
once we have the legal material (namely, the norms used by the Judge and the facts), and if
136 See Section 9.3.2 in this Chapter.
Modally Indiferent Substrate
Dialogical Approach to Logic
127
we then apply the (dia)logical approach to logic, it turns out that it’s not only possible to deal
with these two separated domains, but it’s also possible to justify all the legal procedures by
means of an argumentative process (as it is made in legal science).
This purpose does not aim towards a logical treatment of norms regarding the normative
creation via logic, but it aims, rather, towards the logical analysis of the already created norms,
i.e. of the sentences about their validity.
9.2 The Dialogical Approach to Logic
The dialogical approach to logic was first presented in the end of 1950’s by Paul Lorenzen, and
further developed by Kuno Lorenz137. During the 1990’s, this kind of dialog was adapted and
applied to different classical and non-classical logics138.
The main idea of this approach is inspired by Wittgenstein’s maxim of the “meaning as use”.
Following this idea, one can define a logical constant by its use made by two protagonists
through an argumentative procedure. According to this argumentative procedure, a player
proposes a thesis which he will have to defend against the other player, the latter being
opposed to the thesis by trying to conceive a counter-argument. These two protagonists use
the same rules, with one exception: the formal restriction139. This rule forbids Proponent to
utter an atomic formula (i.e. a formula which does not contain logical connectives) if
Opponent has not already uttered it. This restriction, in spite of its asymmetry, assures the
logical truth, if Proponent manages to defend himself from anyone of his adversary’s
attacks140. This rule guarantees that Proponent’s use of an atomic formula depends on a
justification procedure, that is, the use of this atomic formula is independent of the notion of
truth or falsity. Consequently, if Proponent wins141, the thesis is a logical truth with respect to
the dialogical system used. Moreover, during a dialog, a player can only attack the logical
137 See Lorenzen and Lorenz (1978). 138 This approach allows the combination and analysis of different logics in one and the same frame. See Rahman, S. (2006) for examples or Keiff, L. (2009) for a general view of these different developments. 139 The overall set of used rules settles the dialog system. See. Appendix for a presentation of the standard rules of the dialogical approach to logic. 140 Through the notion of winning strategy. 141 See Winning Rule SR-3, Appendix, § Structural Rules for the definition of the principle regulating victory.
128
structure of his adversary’s discourse, never the content of the discourse, nor even the
adversary himself. If in this form of dialog the players commit themselves to defend their
arguments announced during the play, those commitments are not about the content of the
argument, but about their logical form. In other words, Opponent and Proponent are only
committed to defend the logical structure of the propositions they use.
From a technical point of view, dialogic is not a specific logic, but rather a conceptual
framework providing the study of different logics under an argumentative process. Dialogical
language is defined from a determined logic where the letters O and P, standing for Opponent
and Proponent, are added as well as the symbols “?” and “!”. These symbols are required in
the formulation of the rules of use of the logical constants: the particle rules. Another set of
rules determine the conditions under which the particle rules have been or can be used. These
are called the structural rules.
- Modal Logic
Modal Dialog results from an original idea coming from the researches of S. Rahman and H.
Rückert142. This kind of dialog contextualizes the notion of proposition. If the standard dialog
allows us to define the meaning of the logical constants through their use in an argumentative
process, standard modal dialog allows us to associate a contextual dimension to this notion
of use. The meaning of a logical constant is directly dependent on its contextual use. Not only
the meaning is defined by its use, but this use is also dependent on a context. The context is
then a constitutive part of the meaning of the constant. Moreover, the pragmatic character
of the dialogic semantics displays different interpretations of the modality simply by changing
the associated structural rule. To our purposes, the modal operator receives a deontic
interpretation, that is, we interpret the modal operator Oϕ as “it is obligatory that ϕ” or “ϕ
ought to be the case”.
9.2.1 Dialog, Validity, Truth and Justification
142 See Rahman and Rückert (1999).
129
In this section we will present the formal details of our dialogs. But before that, we must
explore both the advantages and inconvenient aspects of the dialogical approach concerning
our problem, because if the dialogical frame allows an argumentative approach to logic, this
frame is more concerned with the formal truth of a formula than its truth itself. This is the
reason why we need some modifications of the standard dialogical framework.
The next paragraphs lay the groundwork for our conceptual modifications.
- Material Dialog and Justification
The dialogical approach to logic deals with the logical validity of a formula and not its truth
value. But legal science, the discourse about the law, is evaluated in terms of truth rather than
logical or legal validity. Besides, different approaches of the dialogic allow characterizing the
truth value of a formula in material dialog143.
First of all, we need to consider initial concessions, i.e. some hypothesis over which the dialog
will be constructed. Those hypotheses can be used by Proponent to justify his utterance. Thus,
while in the dialogical approach we only consider Proponent’s justification via the procedural
use of atomic formulas, that is, we consider his saying: “I use this formula in the same context
and in the same way you did”, with material dialogs he can justify a proposition saying: “I use
this proposition because it is an initial concession/premise”. This tends to weaken the formal
restriction: Proponent is no longer restricted by this rule, he can use the initial concessions to
utter and/or justify his use of atomic propositions although Opponent has not already uttered
them before.
- Dialog about Truth
From a logical point of view, material dialog deviates from the notion of winning strategy —
which is, in dialogic, the counterpart notion of validity. With the material dialogs, a winning
strategy for a given thesis becomes relative to a set of premises. It’s only with regard to the
set of premises that the winning strategy exists. So, it’s possible to show that the thesis holds
143 See Rahman and Tulenheimo (2009).
130
with respect to the premises given before the dialog, and this fits perfectly with the Kelsenian
purpose saying that the legal theory is a true/false discourse raised from law.
Thus, even if this kind of dialog keeps us apart from the notion of logical validity, we stick to
Hans Kelsen conception that the legal science only deals with truth.
- Validity and Truth?
According to Hans Kelsen, legal science – the discourse about the law – has to be supported
by the law itself, i.e. a norm or a set of norms144. In the same way, the two players of the
dialog discoursing about the law (about the normative creation or the legal decisions) have to
support their arguments over this normative set and over what the Judge qualifies as facts
according to a determined case145.
In a given logical system a rule is not true nor false, but valid. A rule is valid inasmuch as it
holds independently of its application. However, the particular application of this rule can be
true or false, because it requires certain conditions to be fulfilled. If these conditions are not
fulfilled, the rule cannot be applied, but this does not mean that the rule is invalid or false.
The rule remains valid even when it’s not the case that it has been or can be applied in the
over mentioned circumstances146.
In fact, the distinction between logical rule and particular proposition is analogous to the
distinction between valid legal norm and contingent legal proposition. So, this distinction can
be maintained and reconstructed if we distinguish two sets of premises: a set of general valid
rules and a set of contingent propositions, each one of those representing, respectively, the
set of norms and the set of facts.
- Validity and Truth: a Difference concerning Justification.
144 The “Civil Code” or the “Constitution”. See Kelsen, H. (1979, pp. 61-62 and pp. 81-82). 145 Consequently, the corresponding dialog must beforehand suppose as admitted both the set of norms and a set of facts. This fits perfectly with the notion of initial concessions just mentioned. 146 We will come back later to the discussion over validity and truth, with the justification notion.
131
Previously, we have put in evidence a crucial aspect, namely the incidence that a set of
premises has in respect to the question of the justification of the atomic proposition. Standard
dialogic limits itself to a procedural approach of the justification. It means that Proponent’s
use of an atomic proposition depends on a previous use of this proposition by Opponent.
Implicitly, Proponent can justify its use by copying the procedure that Opponent could
possibly follow to justify this proposition147.
Material dialogs allow a justification for the use of an atomic proposition, but not due to its
previous use by Opponent. Rather, the justification can be made because the proposition
belongs to the set of initial concessions. Because of this authorization for this kind of
justification, material dialogs are closer to the notion of truth and farther from the notion of
validity. However, nothing prevents Proponent to use the procedural form of justification to
justify an atomic proposition which do not belong to the set of initial concessions. In fact, it is
possible to remove the formal restriction and to distinguish two distinct forms of justification:
the procedural justification and the propositional justification. On one hand, the propositional
justification allows to produce a justification with respect to truth, and on the other hand, the
procedural justification has the same effects of the formal restriction and consequently
preserves the (logical) validity level. The combination of these two kinds of justification in a
same frame allows us to deal with the truth of a particular discourse and with the validity of
a rule, without reducing the first to the latter and vice versa.
9.3 The (Dia)logical Tools
We have chosen to benefit from the dialogical frame in order to maintain the distinction
between law and legal science. But before developing these formal details, we have to specify
first the dialogical modal frame under which our reflections are elaborated.
9.3.1 Preliminary Notions
With respect to standard propositional dialogic (presented in the Appendix), we have to add
two operators in order to obtain a dynamic modal dialogic. The first operator corresponds to
147 Thanks to the justification rule that we introduce, this procedure becomes explicit.
132
the modal operator O (previously mentioned in the Section 10.2, § Modal Dialogic). The
second operator is the dynamic operator [ϕ ]ψ148 . We interpret this operator as: “if ϕ is
proved, then ψ holds”149. Whereas the introduction of a modal operator in a dialogical
language demands the contextualization of the players’ utterances, the dynamic operator
prefixes the chosen contexts through a list A.
- The Modal Operator O
A contextual point is a positive integer labeling a formula150. The particle rule of the modal
operator, presented in the table below, allows determining who between the challenger or
the defendant will support the burden of the choice of the contextual point. With respect to
our modal operator, the burden of the choice amounts to the challenger. Intuitively, this rule
captures the following exchange: “if the player X utters, in the contextual point i, it is
obligatory that ϕ , the player Y can choose a contextual point j in which X must defend that ϕ
holds.” The structural rule SR-O, displayed below, defines the conditions for the choice of the
contextual point.
The modal Operator O
Structural rule SR-O:
The Dynamic Operator [ϕ ]ψ
148 Commonly known as the Public Announcement operator in Dynamic Epistemic Logic (DEL). See van Ditmarsch et al. (2008) for a DEL overview and Magnier, S. (2013, chap. 6-8), for more details about the juridical use of this dynamic operator. 149 The dynamic operator entails a conditional form, but this conditional form is far from the well- known material conditional. The consequent requires that the antecedent is true. If the antecedent is not true, it cannot be announced and the consequent cannot be evaluated in the submodel where the antecedent was true before its announcement. 150 A contextual point is not an atomic formula but it receives the same restriction, P only being capable of re-use those introduced by O.
Burden and/or object of choice
X-Utterance Y-Challenge X-Defence
O, the challenger chooses a contextual point il
A |i :Oϕ
A |i : ?il
A |i.il : ϕ
To challenge a move (O − A |i...il : Oϕ ), P can choose any contextual point il already chosen by O.
133
In the particle rule as well as for the structural rule of the operator O, the contextual point i is
prefixed by the list A. The necessity of this list appears in the particle rule of the dynamic
operator and its associated structural rule.151
The list A comes to enrich the notion of contextual point. Strictly speaking, it is not a new
contextual point, it remains the same but prefixed by a list of ordered set of formulas. The list
allows keeping track of the formulas coming from the dynamic operator152.
Burden and/or object of choice
X-Utterance Y-Challenge X-Defence
[ϕ ] ψ , the defender has the choice
A |i : [ϕ ]ψ
A |i : ?[ ]
A |i : ¬ϕ or
A • ϕ |i : ψ
Because of the “if. . . then” form of this dynamic operator, the burden of the choice is
supported by the defender. When the player Y challenges the evidence ϕ, X can reply “ϕ is
not the case” or “because of the evidence ϕ , ψ is the case”153.
The structural rule SR-A specifies certain conditions for the use of the list. The general idea of
this rule is to ensure that a player who adds a formula in the list A is able to justify this
formula154.
Structural rule SR-A:
151 These rules are originally introduced in Magnier, S. (2011); soundness and completeness proof of them is given in Magnier and de Lima (2014). 152 See Magnier, S. (2013, chap. 5). 153 The dynamic operator (ϕ )ψ , such that (ϕ )ψ def ¬[ϕ ]¬ψ , expresses the same idea but with a conjunctive form. So, in this case the burden of the choice is not supported by the defender but by the challenger. 154 This rule is the counterpart of the fact that only true formulas can be used with this dynamic operator—what can sound a bit idealized if we consider evidence. An interesting point would be to consider refutable evidence, but to do this we need to use more sophisticated dynamic operators.
For any move (X − ϕ1 ...ϕn |i : e), player Y can compel X to utter the last element ϕn of the list A :
• in the contextual point i : (Y − ϕ1 ...ϕn−1 |i : !(ϕn ) ) or
• in the contextual point j : (Y − ϕ1 ...ϕn−1 | j : !(ϕn ) ) if e =? j .
134
The rules presented above generate a dynamic modal dialogical frame. This dialogical frame
allows evaluating the logical truth of modal dynamic formulas.
9.3.2 Set of F Facts and the Problem of Justification
In order to reach the truth level, we have to introduce the sets of premises called Norms and
Facts155 in the dialogical frame obtained, and re-found the question of justification of atomic
formulas.
Definition 1 (Fact) We take as a fact all data accepted by the Judge. These data are expressed
by formulas in the set called F . This set is fixed before the dialog starts, and these formulas
can be used in any contextual point of the dialog.
Those facts represent the data accepted by the Judge. For instance, in our example, “Jérôme
C. has committed tax evasion” or “Jérôme C. has not committed tax evasion” are facts.
Recall: the two players can utter atomic formulas because we have eliminated the formal
restriction. However, in order to maintain an internal coherence of the dialog, if a player uses
an atomic formula, his adversary can compel him to justify it. That’s why a justification rule is
required.
The justification rule (PR-J) allows distinguishing between two different kinds of justification:
1. Propositional justification, and
2. Procedural justification156.
155 As discussed in Section “Validity and Truth?” 156 Even if we present it in two distinct tables, it is in fact one and the same rule authorizing two different defenses.
135
This differentiates a justification which is based on the set of facts, from another which is
founded in the argumentative process itself. It’s possible for the latter kind of justification to
not even be based on the facts157. A consequence of this is the fact that we need to add the
following logical constant in our logical language:
[p] for all atomic propositions,
where [p] is a justification for the atomic proposition p.
- 1. Propositional Justification
A propositional justification can only be produced according to the set of facts F, i.e. an atomic
proposition can only be propositionally justified if it belongs to the set F . This rule is described
as the following:
Propositional justification X-Utterance Y-Challenge X-Defence
A |i : p, the challenger requires
a justification for proposition p.
tion p.
A |i : p
A |i : ![ p ] A |i:
[p] ∈ F if p ∈F
The set of facts (F) and the rule for the propositional justification allow establishing a perfect
symmetry between Proponent and Opponent with respect to the moves they can make in the
dialog. So, the two players have to use strictly the same rules158.
- 2. Procedural Justification
In order to fully understand the rule of procedural justification, let us suppose that, during the
play, player Y introduces an atomic proposition and, then, player X utters this same atomic
proposition. Player Y can compel X to provide a justification for this atom. In this case, X can
defend himself by copying Y’s (possible) justification. That is possible only if this proposition
157 We come back to this distinction in the paragraph “Propositional and Procedural Justification”. 158 Normally, the two players do use the same rules, but the formal restriction introduces an asymmetry. Our justification rule allows establishing a strict and complete symmetry
136
belongs to Y’s moves in the play, what we note as Y-moves (MY ) in the rule displayed in the
following table159:
Procedural justification X-Utterance Y-Challenge X-Defence
A |i : p, the challenger requires a justification for the proposi- tion p.
A |i : p
A |i : ![ p ]
]
A |i : [p] ∈ MY if
{ A |i : p} ∈ Y-move
The procedural justification consists in justifying the uses that a player makes of an atomic
proposition by a previous use made by his adversary in the game. Thus, when a player use this
kind of justification, he affirms nothing more than “it’s legitimate for me to use a proposition
in the same conditions as you did”.
- Propositional and Procedural Justification
While the propositional justification allows a link between an atomic proposition used in the
dialog and its truth value fixed in the set of facts, the procedural justification allows us to use
propositions that do not necessarily belong to the set of facts (F). The conditions of use of
these propositions are identical for the two players. During the game, if a player introduces a
new proposition, he authorizes his adversary to use this proposition under the same
conditions and according to the same justification. Concerning the argumentative aspect, the
main focus is not over the fact that a proposition belongs to the set F (consequently, over its
truth value), but over the symmetry of the conditions of use of these propositions, i.e.
whatever X does, Y can do the same.
Moreover, this strict symmetry between the two players let us consider the two players as
having the same cognitive capacities (they can make the same inferences) and identical
argumentative means (there is never imperfect or hidden information). From a strict juridical
point of view, it would be difficult to consider an argumentative debate over law where the
two protagonists won’t have the same argumentative tools.
159 This rule gives a typical copycat procedure, i.e. a procedure identical to the formal restriction.
137
- Factual Truth & Formal Truth
The choice with respect to the nature of justification—propositional or procedural— is crucial
because it permits to distinguish two levels of truth: factual truth and for- mal truth. On one
hand, the propositional justification is based on the link between the propositions uttered
during the dialog and the facts (admitted in F). Thus, this kind of justification allows us to
establish a correspondence between the discourse and the facts, which is precisely the
definition of truth. On the other hand, the procedural justification doesn’t focus on the truth
of discourse relatively to the facts.
Procedural justification is an internal form of justification, independent from the truth of the
facts. It’s based on the formal truth of the propositions, on the propositions’ structure
independently of their relation to the facts. This reveals a difference between the true
propositions with respect to the facts, and the justifiable propositions with respect to a given
procedure. If an atomic proposition belonging to the set F is necessarily true and can always
be justified, the same doesn’t happen with a justified atomic proposition, which is not
therefore necessarily true. The procedural justification authorizes a justification
independently of the propositional or factual truth.
The notion of justification is then founded over a larger notion than that of truth. It is founded
over a player’s ability to defend a proposition, so that a player can defend himself without
using the truth notion. It’s the player’s use of the justification rule which finally determines
the discourse level.
9.3.3 The N Normative System
Definition 2 (Norm) For “Norm” we take the law used by a Judge and recognized as so by the
two protagonists of the dialog. Those norms are formulated by general rules and constitute
the set of norms N. This set is fixed before the beginning of the dialog and the general rules
can be used in any contextual point of this dialog160.
160 The set N cannot contain conflicting norms. This aspect is somewhat idealized, because it is not impossible for the norms to enter in conflict. The question concerning conflicts between norms and their choice is not of our interest by now, but could certainly be the object of future researches.
138
A particular norm is produced from a more general one. This process passes by the Judge, who
has a crucial role in the creation of the particular norm. He needs, on the one hand, to be
based in the existence of a general norm with a corresponding content and, on the other hand,
the particular fact concerning the affair in question. From that, once authorized by law itself,
he is able to create a particular norm. According to Kelsen, a legal authority can fill up this role
of creation of a particular norm. In the kind of dialogs we deal with, those general norms are
reconstructed by the logical rules which determine the set N, i.e. the logical set of general
rules should be seen as the counterpart of the legal set of general norms. These norms can be
used by the players during the dialog, but under certain conditions, which are specified in the
following paragraphs. Before that, we have to mention a particularity concerning the use of
the legal order in a dialog.
When a player wants to use one of the general norms of the set N, he expresses a question
about the use of this general norm (which was used or could be used by the Judge). It cannot
be a challenge against a previous move because players are not allowed to create particular
norms, only the Judge is allowed to do this. So, the rule must be read in the following manner:
“from a general norm belonging to the legal order, any player can ask his adversary if the
Judge has used or could have use this norm to take his decision over an individual, which is
represented in the rule by the choice of the individual constant a.”
Rule for N
N Y-Question X-Defence
(Ax → OBx) ∈ N , Y asks if the norm can be applied to the agent a.
A |i : Ax → OBx
A |i : ?a/x
A |i : Aa → OBa
In the formulation of the rule for the use of N , the predicates A and B are used to, respectively,
designate an Act made by an agent x and a behavior to be imputed to the same agent x
concerning this act. The “Ought” of this behavior is translated by the operator O presented in
the Section 9.3.1.
- N Normative System & its Conditions of Use
139
If Y, in his question about the set N, can choose an individual constant, he is not therefore
allowed to arbitrarily introduce a new individual constant in the play. The individual constant
chosen must be previously given, i.e. at least one occurrence of the individual constant must
occur in the play.
But the simple occurrence of the constant is not yet sufficient for the player to be able to
request the set of norms. The antecedent of the demanded particular norm, containing the
individual constant in question, must be a justified element in the list A. Thus, the particular
norm Aa → OBa can only be obtained during the play if the defense for [Aa]ψ leads the player
to add Aa to the list A , that is, if there is an evidence that a satisfies the antecedent of the
norm.
In fact, every justified proposition added to the list is called an evidence. But, since a
proposition in the list A can be justified simply by its belonging to the set of facts F or, for
those which do not belong, by the procedural justification, two kinds of evidence need to be
distinguished:
1. A ∩ F ; and
2. A \ F .
The set A ∩ F describes the set of propositions belonging to the list A and to the set of facts,
while A \ F delimits the set of propositions belonging to the list A minus the propositions
belonging to the set of facts. The justification of a proposition of the list by the set of facts (1)
ensures that the demanded particular legal norm is created by the Judge with respect to the
facts established and accepted by both players.
So, the Judge can create the particular norm because it is established that the act committed
by a is a fact. If a general norm imputes a behavior B according to an act A, and if it’s observed
that, in the facts, the individual a has committed the act A, the Judge can create the particular
norm imputing the behavior B for individual a.
However, for the propositions which do not belong to the set of facts (2), the justification can
only be procedural. The subsequent evidence is not based on a veridical discourse (held by
the notion of truth), but on the ability to defend by using the adversary’s arguments. This
warrants the validity of the Judge’s created norm, independently of his having recourse to the
factual truth.
140
- What is and What Could Have Been…
The propositions in the list A depend on the players’ ability to justify them. If we consider that
the set F attributes a truth value to propositions belonging to it, propositions added to the list
and justified using this set (propositional justification) describes what is, whereas propositions
added to the list using the procedural justification represent what could have been or could
be the case, but which is not the case with respect to the facts. From this, we can go on and
distinguish two levels of discussion between Opponent and Proponent:
1. For all propositions verifying A ∩ F, the players discuss about the particular norms created
by the Judge with the established facts—the truth concerning the application of the particular
norm with respect to the established facts.
2. For all propositions verifying A \ F, the players discuss about the particular norm that the
Judge could have created or could create if the antecedent of the norm were/becomes an
established fact—the validity of the general norm via the possible particularization
independently of the facts.
From a strictly grammatical point of view, in the second level of the discussion, the question
posed about the particular norm created by the Judge changes its mode. The first level of
discourse corresponds to the indicative tense: there is a norm created by the Judge. The
second level of discourse is performed under the past or present conditional tense: he could
or could have created a norm such and such. . .
This changing of mode is interesting because it puts in evidence the conditional aspect of the
normative creation regarding the existence of the fact. It is true that the general norm is
formulated under a conditional proposition, but its use is itself conditioned by the existence
of a fact corresponding to the normative antecedent. If the antecedent belongs to the set of
facts, the grammatical mode is the indicative tense: Opponent and Proponent start the debate
from the existence of a particular norm whose creation was linked to a fact. But, if the
antecedent belongs to the list of announcements, but not to the set of facts, then the
grammatical mode is the conditional tense. Opponent and Proponent consider the particular
norm that could be or could have been created if the existence of the fact was or could
someday be recognized. The procedure to follow is identical no matter whether the fact exists
or not: the difference lies in the level of the mode and the justification. To the existing facts,
it is the propositional justification which takes place (or the procedural justification, if the
141
propositional justification was already used). To the non-existing facts, only the procedural
justification can take place. So, the focus is not over the existence of the fact, but over the
players’ ability to justify their use of those facts. The players do “as if” the fact existed, even if
it is not actually the case. In that case the fact is only used as an element inside a procedure.
The logical formalization which we introduced allows Opponent and Proponent to discourse
about the particular norms created by the Judge, according to the facts and to the norms that
could or could have been created if the existence of the fact was proven. In order to emphasize
this distinction, we distinguish in the list what is justified by the set of facts from what can
only be justified by a procedural means. For every proposition added to the list without being
propositional justified (i.e. all proposition ϕ ∈ {A \ F }), we add ∗, such that ϕ ∗ means ϕ can
only be procedurally justified. Thus, in a play, the propositions ϕ1 , ...ϕn in the list A have a
truth value, while the propositions ϕ ∗, ...ϕ ∗ have a value determined in terms of their
capacity of being defended, i.e. a formal value.
9.4 The Dialogical System DLLC2
We define the Dialogical Logic for Legal Condition DLLC2:
• Particle Rules = PR-SC ∪ PR-MO ∪ PR-DO ∪ PR-J ;
• Structural Rules = SR-0 ∪ SR-1 ∪ SR-3 ∪ SR-O ∪ SR-A
These sets of rules must be used in a restrict dialogical frame, i.e. determined by the sets F
and N. These two sets determine the conditions for the material dialog DM such that DM = F
∪ N .
Definition 3 (DLLC2 ) DLLC2 is defined by the union of sets Particle Rules and Structural Rules
used in DM:
DLLC2 = PartRules ∪ StrucRules
DM
142
9.5 Jérôme C.’s Guilt and Further Discussions
Since the formal details are now presented, we can return to the example of the previous
chapter and use those formal detail to explore specifically our Jérôme C. example.
Subsequently, we will discuss some properties of our dialogical frame underlined by the
example.
9.5.1 Jérôme C.’s Example
In our Jérôme C.’s example we have used the norm “Tax evasion must be punished”. We
reconstruct this norm via the logical rule Tx → OPx . Whereas the set N doesn’t change, we do
change the set F. Let us for now assume that “Jérôme C. has committed tax evasion” is a fact
admitted by the Judge, as Tc (table below), and then let us consider that Tc is not admitted by
the Judge and therefore does not belong to the set F (two following dialogs). As we will see,
when Tc does not belong to the set F , the distinction between the two modes of discourse
arises.
Initial concessions for the dialog below
O P ε |1 : [Tc ]OPc 0
m := 1 n := 2 1 ε |1: ?[ ] 0 Tc |1 : OCa 2
3 ε |1 : !(Ta ) 2 Tc |1 : Pc 4
5 ε |1 : ! [Ta ] 4 Tc |1 : [ c 6
7 Tc |1 : ?2 2 Tc |1.2 : Pc 14
9 Tc |1 : Tc → OPc N Tc |1 : ?c/x 8
11 Tc |1 : OPc 9 Tc |1 : Pc 10
13 Tc |1.2 : Pc 11 Tc |1 : ?2 12
Jérôme C. commits tax evasion
N
TX → OPX
F
TC
143
In the dialog above, Proponent adds Tc in the list A in move 2. In the next move, Opponent
uses the structural rule SR-A to compel Proponent to utter this proposition in the contextual
point 1, what he does in move 4. Then, Opponent requires a justification for the proposition
Tc . Proponent manages to justify himself without difficulty, since Tc belongs to the set of facts
(move 6). Once Tc is a fact belonging to the list of announcements, Proponent can request the
legal order by asking for the particular norm created by the Judge with respect to fact Tc (move
8). Accordingly to this particular norm, Proponent manages to defend that it is true that
Jérôme C. has committed tax evasion, and then it is obligatory that he has to be punished
(move 14).
Now we consider what would happen if Tc is not an admitted fact.
Initial concessions for the next two dialogs below
O P ε |1: [Tc ]OPc 0
m := 1 n := 2 1 ε |1: ?[ ] 0 ε |1: ¬Tc 2
3 ε |1: Tc 2 – 3 ε |1 : ![Tc ] 4
Choice 1 – Indicative Tense
In the dialog above, Proponent chooses not to commit himself with the defense of Tc and he
defends with ¬Aa. Opponent can then only challenge the negation (moves 2 – 3).
After this Opponent’s challenge, Proponent has the choice:
1. He can require for a justification of Tc , or
2. He can use Opponent’s challenge to change his previous defense.
N
TX → OPX
F
Ø
144
In the table above we suppose that Proponent makes the choice 1. We develop the possibility
of the choice 2. in the next dialog below.
Choice 1: After Opponent’s move, Proponent requires a justification for Tc (move 4). Since Tc
does not belong to the set F, Opponent cannot use the propositional justification.
Unfortunately for Opponent, Tc does not belong to any Proponent’s previous moves. So, he
cannot use the procedural justification neither. Consequently, Opponent loses the play.
Without proving Jérôme C.’s culpability, no particular norm could or could have been created
by the Judge.
Choice 2: Opponent’s challenge in move 3 allows Proponent to change his defense and then
add Tc in the list A (move 4). Opponent uses the rule SR-A to compel Proponent to utter Tc in
the contextual point 1 (moves 5-6). In the next move he requires a justification for Tc.
Proponent cannot use the propositional justification, since Tc does not belong to F.
Nevertheless (contrarily to the table considering Choice 1), the proposition Tc belongs to a
previous move of Opponent. He had already uttered Tc when he challenged the negation
(move 3). Regarding this move, Proponent has not required a justification, so he confers
himself the possibility to reuse this proposition under the same conditions, and to justify it
through a procedural justification. After this procedural justification, Tc gets marked by a ∗ in
the list, manifesting the change of mode of the discourse. The progress of the play is then
based on what would or could have been the case once Tc was an accepted fact. In move 10,
Proponent asks for the particular norm that the Jude would or could have created if Tc was a
fact. The rest of the play is similar to what is developed in the dialogue prior to this one, with
the exception of the value of the players’ discourse: they no longer debate over what is true,
but rather over what could be or could have been true. The discourse is developed
independently of the facts. It deals with the general norm’s validity once its possible
particularization, doing as if the fact was a verified and accepted one.
O P ε |1 : [Tc ]OPc 0
m := 1 n := 2 1 ε |1: ?[ ] 0 ε |1 : ¬Tc 2
3 ε |1 : Tc 2 ⊗ Tc |1 : OPc 4
5 ε |1 : !(Tc ) 4 Tc |1 : Tc 6
7 Tc |1 : ![Tc ] 6 Tc |1 : [ ] ∈ C Tc O 8
9 T∗c |1 : ?2 4 T∗
c |1.2 : Pc 16
11 T∗c |1 : Tc → OPc N T∗
c |1 : ?c/x 10
13 T∗c |1 : OPc 11 T∗
c |1 : Tc 12
15 T∗c |1.2 : Pc 13 T∗
c |1 : ?2 14
Choice 2 – Past or Present Conditional Tense
145
9.5.2 Further Discussions
The comparison between choices 1 and 2 has an important didactic aspect. Let us consider
the object of the discussion between Proponent and Opponent:
Proponent presents the following argument: “if it is proven that Jérôme C. committed tax
evasion, then the equivalent sanction ought to be applied”. But, before entering in this
discussion, Opponent and Proponent agree to recognize that it is not yet admitted that
Jérôme C. committed tax evasion. Consequently, Proponent only reaffirms that if Jérôme C.
has committed tax evasion (which he is or could have been the suspect), it would be obligatory
that he ought to be punished. The fact of his culpability does not change the fact of the
obligation of the sanction in the case of culpability. Based on F, Proponent shows that it
cannot be obligatory that Jérôme C. is punished if it is not established that he is guilty of tax
evasion (table considering Choice 1). Using the procedural justification, it’s reaffirmed the
dependence between “being obligatory” and the satisfaction of the condition (i.e. the validity
of the general norm). This aspect manifests the imputation aspect linking the condition (have
committed tax evasion) to the “ought” element (the sanction: to punish the behavior). If the
condition is not fulfilled, the “ought” concerning the behavior cannot take place. The tax
evasion is the condition for the “ought”, for the sanction, but the sanction is conditional
because it rises from a modality. If the latter is not itself fulfilled, Proponent wins, but only
because the conditional relation is not invalidated by the non-satisfaction of the condition.
Thus, the norm remains valid independently of the fact that Jérôme C. has not committed tax
evasion.
- Back to the Evidence of the Illegal Act
If the “ought” is associated with a condition, this condition has a particular character. It
consists in the satisfaction of the particular norm’s antecedent, what ensures the formulas
added to the list A. Remember that these formulas have a particular status. When a player
changes the situation of the argumentative process (by choosing a new contextual point), due
to the rule SR-A his adversary can compel him to (re)utter and then (re)justify them in this
146
chosen situation. Consequently, since a formula is introduced to the list A, a justification for
this formula can be required in any new contextual point. When players are incapable of
producing this justification, the particular norm cannot be obtained in this contextual point.
Hence the punishment could not be said to be obligatory in it.
In order to achieve the sanction, the guilt has not only to be a fact, but especially it must be
proved, that’s to say justified in any situation. It could not exist a situation in which the
sanction is pronounced and where the guilt cannot be justified. Jérôme C. cannot be punished
for a tax evasion that he has not really committed. But this doesn’t mean that he should never
be punished at all: only that, while in absence of a proof of his guilt, no sanction can be applied.
Even if everybody knows that he is guilty, that’s not sufficient for the sanction to have to be
applied. It must be proven that Jérôme C. is guilty, and the formalism surrounding the dynamic
operator explicitly allows this. Moreover, it remains possible for Jérôme C. to commit tax
evasion without becoming a suspect of such an act. For this, it’s enough that we cannot justify
this fact in any situation.
For a sanction to be applied to a guilty behavior, we need that an authorized person (the
Judge) creates a particular norm from a more general one, but also that this behavior is
proven161.
Remark: Deontic logic, in its treatment of legal norms, considers them as instantiations of
general norms. The problem with this approach is that it reduces the normative propositions
to conditional instantiations162 , what addresses deontic logic to paradoxes linked to the
material conditional. In our reconstruction of the legal reasoning, our proposal also deals with
the (material) conditional to formalize the general norms. But the use of those norms
presupposes the creation of a particular norm (to be created by the Judge only).
The use of this particular norm is itself dependent on the evidence of the antecedent, which
can only be attained by the dynamic operator. Thus, the obligation of the consequent is
required only when the antecedent is true or procedurally justifiable. Consequently, due to
this formalization, not only the falsity of the antecedent cannot lead to the trivialization of the
conditional relation, but also it allows us to study the meaning of a norm through its
161 Even if the set F contains ¬Tc or doesn’t even contain Tc, identical plays will be produced. Maybe this indicates a link between the absence of the guilty proof and the innocence presumption. 162 This is precisely the context where Jørgensen’s dilemma emerges.
147
description, in order to give the conditions of its use when the fact is missing (procedural
justification) or to give the truth conditions of the particular norm if the fact is the case
(propositional justification).
9.6. Back to Jørgensen’s Dilemma and Final Remarks
This original dialogical approach to logic is the most relevant frame to display the process
concerning normative creation through an argumentative practice. We respect and preserve
all the Kelsenian theses with respect to the dualisms between the realm of “Is” and the realm
of “Ought”, between law and its science or theory, between a norm and its description,
between the “legal actor”—the Judge, the Legislator, the authorized person—and the
persons external to the legal procedures—the jurist, the legal scientist. All those elements
play a specific role in our study of the legal reasoning through the process of a dialog, and they
are essential to show how mistaken is the notion of a practical syllogism, which lead to
theoretical problems like Jørgensen’s dilemma.
Regarding the dilemma, Kelsen had pointed out the fact that to create a particular norm (in
the conclusion of Jorgensen’s practical syllogism) the creator has to have the power to do so.
In the legal context, this power, this authorization, must come from the law. Legal norms are
always created via procedures which are internal to law itself. No one from the outside is able
to create a valid norm, even with the support of the existence of a more general norm with
the correspondent content. This is a necessary, but not sufficient, step in the normative
creation. The “authorization” aspect cannot be ignored, and can never be attained by
someone who is outside from the legal sphere. Each one of us, as citizens, is able to put a
norm in question, to evaluate it as fair or unfair, to consider if it should be or not applied to a
determined case. But Kelsen emphasizes that those are all acts of thought that do not attain
the legal level of the valid act of will, which could be produced by anyone but the Judge, or
the authorized person.
With the dialogs, what happens is a procedure of justification after this act of will has taken
place in time. Moreover, it is a procedure which allows considering what would be the case
once other conditions were at stake. Proponent and Opponent have no legal authority, they
only discuss about the legal procedure, once it has already been achieved. Certainly, this can
148
clear up a lot of legal aspects such as mistakes or flaws to be revealed in the process, and that
is all the interest of a legal theory.
This approach fits perfectly with Kelsen’s solution to Jørgensen’s dilemma, saying that we are
all capable of questioning, justifying and reasoning about norms.
We can guess what would be the result (the conclusion on the misleading practical reasoning),
i.e. what would “normally” be the content of the result of a legal decision, for example. But
by no means are we able to state this “result” as being a valid norm. There is no way to
trespass the inference barrier in the practical syllogism, because there is no such a way for a
syllogism to be a practical one: syllogisms are constructed by acts of thought, and practical
decisions are made by acts of will. It is a mistake to accept that we can simply mix up the two,
or translate one into another, as it is a mistake to consider that norms can result of acts of
thinking.
Finally, we believe that our dialogic approach shows a perfect example of an application of
Kelsen’s theory concerning the important question of the normative creation. It displays how
it is possible to confront oneself to general norms like “Tax evasion must be punished” and
particular norms like “Jérôme C. (who committed tax evasion) must be punished” from a
completely external point of view, and the justification element is central in this undertaking.
We preserve Kelsen’s temperance about the modally indifferent substrate, yet showing how
far we can still go in analyzing the different possibilities of interpretation. We show that it is
perfectly acceptable to analyze, theorize, argue, examine, debate about norms, without
transgressing the limits between the normative (practical) and the descriptive (theoretical)
levels, and still preserving the methodological principle of “purity”. From this conscient and
solid methodological approach, no harming puzzles arise.
149
Chapter 10
Conclusion
The present study represents a real effort towards testing the limits of approximation
between the fields of Law and Philosophy. This undertaking is justified by the fact we expect
have made clear that, despite refusing the possibility of apllying logical principles such as the
principle of non-contradiction and the rules of inference directly to the legal field, Kelsen
seems more enthusiastic regarding a dialog between the two disciplines when the notions
coming from the philosophical field serve to the legal science as a means to approach its
objects, namely, the legal norms.
Let us then make a recapitulation of the contents explored in the present work.
As we stated in the Introduction, our purpose with the present work was to investigate the
stability of the possible relations between the fields of Philosophy and Law. Since our context
of analysis is the legal theory proposed by Hans Kelsen, this purpose is undertaked by taking
as background the traditional problem of the possibility of the application of logical principles
to norms.
This application was a widespread practice regarding Kelsen’s contemporary colleagues, while
his view on the issue was always marked by a strong skepticism. Thus, while the general
practice in legal theory seemed to take for granted the identification of legal norms with true
sentences, Kelsen’s writings grew more and more explicit concerning his negative perspective
of the approximation between the fields of Law and Logic.
The originality of our approach consists in showing, as explicitly as possible, that the rejection
of Kelsen concerning the relations between Law and Philosophy is not so straightforward,
even – and, we dare to say: especially – in the last work, the General Theory of Norms, where
150
Kelsen meticulously express by all means the differences between the indicative and the
prescriptive domains.
We defend that Kelsen’s treatment of the problems involving that relation is complex. The
thesis we tried to defend points to the fact that, while the direct application of logical
principles to norms is expressely denied, the use of some methodological principles and
concepts belonging originally to the philosophical field are encouraged for the work of the
legal scientist, in his task of describing, interpreting and justifying the intra-relations of the
normative system. Those concepts are mainly represented by the naturalistic fallacy and the
fiction as a methodological tool, and the philosophy authors praised by Kelsen are
represented, mostly, by Immanuel Kant, David Hume and Hans Vaihinger.
We also analyzed the alternatives that Kelsen offers when rejecting the logical principles, such
as the case of derogation as an alternative to the principle of non-contradiction, and the
notion of modally indifferent substrate as a means to understand the process of normative
creation without making reference to the rule of inference or to structures such as the
misleading practical syllogism.
Other than that, we offered our own attempt to logically formalize the legal procedure of
justification of the normative creation, which we intend to have presented as a legitimate
instantiation of the Kelsenian theses of preservation of the methodological principles
guaranteeing and attesting the “purity” of Law.
In order to better recognize and corroborate each one of the contributions we provide for the
question of the relations between philosophy and the legal field as it is considered by the
positivist tradition, let us now analyze, one by one, each of the chapters composing the
present study.
In chapter 2, “The Author and his Theory”, we tried to provide the conxtext and the motivation
that pushed Hans Kelsen to develop his scientific theory for the study of positive legal orders.
We explained the difference between legal positivism and the alternative schools of thought
like the legal realism and the natural law tradition.
This comparison was useful to make clear why the frame of the positivist tradition was chosen
by Kelsen for his project of a “pure” theory of law, which preserves the autonomy of the
normative field concerning, for example, other disciplines, such as sociology or religious
oriented moral approaches.
151
Concerning Kelsen’s works, we explained the fact of the differences between his approach in
the Pure Theory of Law with respect to his later work, the General Theory of Norms, especially
regarding the question of the possible relations between law and logic. We also gave a general
overview of the academic life of Kelsen, from his first works in Vienna until his later departure
to the United States.
In chapter 3, “Preliminary Notions”, we offer a wide overview of the main notions composing
Kelsen’s special terminology for his legal theory. We start with the notion of legal validity, with
its particular meaning of normative existence. Once knowing that validity equals to the
existence of a norm, and defining the legal norm as the objective meaning of an act of will, we
explained how those norms relate to the science describing them, and the importance of not
falling into the naturalistic fallacy.
Then, we discussed the formulation of the positive norm, and how the norm is not a simple
imperative, because of its binding force. Regarding the normative construction based in
conditional or hypothetical norms, we also remarked the differences between causality and
imputation. Since those notions form a “conceptual web”, we are able to inter-relate them,
noticing that, for example, concerning the dichotomy of causality/imputation, it is in fact
directly linked to the division between science and its object, which is by its turn regulated by
the methodological abyss between “Is” and “Ought”, which should be kept separated from
each other in order to avoid the falling into the naturalistic fallacy…
Still in this rather long chapter, we also took a glimpse at the notion of fiction in the context
of legal positivism. This notion was introduced when we first explained the construction of the
normative system according to a hierarchical structure, having at the top of it the problematic
notion of basic norm as a fiction, regulating and authorizing the creation of the legal norms of
the system
Then, we approached the two main questions dealt by Kelsen in the General Theory of Norms.
The first is the possibility of the application of the principle of non-contradiction in the case of
conflicting norms. Kelsen denies this possiblitity, by making reference to the notion of
derogation as a legitimate legal means to deal with the conflict. The second is the possibility
of the application of the rule of inference in the case of normative creation. This procedure
would be expressed by the practical syllogism, representing the use of the practical resoning
in order to obtain a new norm in the conclusion. Once again, Kelsen will deny this possibility,
152
by attacking all and every comparison or identification of the valid norm with a truth-valued
sentence.
After this general terminological background, we were able to finally approach the specific
questions concerning the relations between law and philosophy, one by one, as well as giving
our contribution to the treatment of the different topics.
So, in chapter 4, “Legal Positivism: A Defense of the Scientific Method”, we tried to collaborate
in the discussion concerning the historical antagonism between the currents of legal
positivism and natural law. After explaining to a greater extent the reasons for the hostily
regarding the defenders of each of of those approaches, we heed the fact that it would still
be perfectly possible to build a scientifical approach to morality as far as it consist in positive
morality.
That remark is based on the division between static and dynamic principles operating in the
intra-normative relations. Actually, the last work of Kelsen intends to be a theory for norms in
general. Besides, most of the examples Kelsen uses to illustrate his thesis, not only in the
Ganeral Theory of Norms, are examples coming from positive religion. We defend the view
that the positivism qua method of knowledge of a given set of norms is capable to report also
norms involving non-scienfical notions in their content, i.e., to attest their validity as posited
norms forming a system ruled by a static principle. This aspect may represent a softening of
the tensions involving the two kinds of approach.
In chapter 5, “Science and Method: The Naturalistic Fallacy”, the relations between the fields
of philosophy and law were finally tighten, when Kelsen recognizes many similarities in David
Hume’s skeptical and above all fundamentally empirical approach to the knowledge of a
determined scientific object. Kelsen recognized the precepts of the legal positivistic
perspective in Hume’s conceptions, especially concerning the method of scientific approach,
which seems to fit perfectly to Kelsen’s “purist” ideal. We analyse and reinforce the closeness
between Hume’s and Keslen’s approaches.
Chapter 6, called “The Non-Existence in Legal Science”, we dealt with another very
philosophical subject, that of non-existence, which was developed and studied by means of
the notion of fiction. Again, as he have done concerning the philosophy of David Hume, Kelsen
grounds his approach in the notion of fiction proposed in the “philosophy of the as-if” of the
German philosopher Hans Vaihinger.
153
Kelsen’s interest is on presenting the notion of basic norm as a fictional methodological device
allowing the comprehension of the legal system as an object of knowledge. We believe to
have offered a clear survey in favour of that perspective, by explaining the purely
methodological use (rather than a prescriptive function) of the notion of fiction as a scientific
tool to deal with the problem of justification, and insisting that the appeals for the fiction is
made only by the jurist in the precise moment of having to face the legal system as a
cognoscible object. Our original approach concerns the attempting to the fact that, in the
same way as the basic norm is not placed in the legal system together with the other norms,
it is also not placed in the scientifical level as a scientific principle. We insisted in the notion of
basic norm as a presupposition during the specific act of the scientist when he is confronted
to the problems of legal grounding.
In chapter 7, “Normative Conflicts and Temporality in Law”, the Kelsenian view concerning the
notions coming from the philosophical (or logical) field and directed to the legal field shifts to
a very negative one. Actually, Kelsen is simply remaining loyal to the idea that logic or
philosophy cannot be applied directly to the legal norms, but only to legal science as a means
to facilitate and help to approach the law. So, when in this chapter we analyze the problem of
conflicts of norms, Kelsen says that the application of the principle of non-contradiction
consists in a methodological error, because that principle presupposes the existence of truth
values, which cannot be attributed to legal norms.
Instead, Kelsen introduced the notion of derogation as a normative function destined to deal
with the cases of normative conflict. In the same way, we have also analyzed the use of
derogation in cases other than the conflict, namely the cases of lost of efficacy over time.
The purpose here was to make explicit the selection concerning the admissions Kelsen seems
to be ready to approve regarding the logical/philosophical elements in the field of law. Since
the case of the application of the principle of non-contradiction represents an attempt to
actually trespass the methodological abyss, by performing the principle’s “function” inside the
normative level, by treating norms as if they had truth values just like the sentences of the
indicative mode, Kelsen is categorical concerning the refutal of any possibility concerning that
application.
The same happens in chapter 8, “The Notion of Practical Reasoning (Part I)”, concerning the
rule of inference in the context of normative creation. The admission of this rule would be
normally expressed in structures such as the practical syllogism, for example, where valid
154
norms and true sentences are identically regarded and treated by traditional bivalent logic.
The point is that, even if most examples of practical inferences seem to be, intuitively,
evidently valid, there is nevertheless no means to legitimate those arguments.
This interesting puzzling aspect was remarked by Jørgensen and named after him by Alf Ross
as the “Jørgensen dilemma”. In this chapter we have studied how Kelsen criticized the answers
that were offered by Jørgensen himself and also, later, by Ross concerning the dilemma. Then
we analyze Kelsen’s own answer to the dilemma, based on the important notion of modally
indifferent substrate.
The notion of modally indifferent substrate can be seen as the main answer Kelsen gives to
the overall attempts in applying logic to norms. An this is because of the fact that this notion
must be understood as being itself the actual content of the legal norm, a neutral element
that can assume either the indicative or the prescriptive forms, but that in itself has no value
whatsoever: it is impartial, indifferent. The consequences on the adoption of such a notion in
the context of legal positivism is the ultimate refutal of any identification between sentences
and norms, since they represent simple the two modes where the modally indifferent
substrate may be inserted, but without having to establish any relation between the modes
themselves.
In this manner, after introducing the notion of modally indifferent substrate, we proceed, in
the chapter 9, called “The Notion of Practical Reasoning (Part II)”, to an attempt towards a
legitimate formalization of the process of normative creation once it has already happened in
time, as well as the justification of that juridical process. In order to succeed in such attempt,
we use the dialogical approach to logic to display the process of normative creation without
infringing either the methodological “abyss” between normative and indicative fields, neither
the principle of “purity” so cherished by Hans Kelsen. More than that, we intend to have
collaborated to the building of solid relations between the legal and the philosophical (or
logical) fields, based on the solid methodological ground that is expected concerning the
Kelsenian approach of legal norms.
This last chapter of the thesis, where we have provided our attempts towards a formalization
of some legal processes without logically “affecting” the normative level, and thereby still
preserving the so appreciated notion of methodological “purity”, might also be a first step
towards further possible developments in the legitimate and conscient approximations
155
between the domains of logic and law. We hope the present work has offered the solid basis
and a clear background for such future undertakings.
156
Appendix
Propositional Dialogic in a Nutshell
This appendix present the rules of Propositional Dialogic, i. e., the particle rules for the
propositional connectors ¬, ∧, ∨, → and the standard structural rules.
How should the particle rules be read?
The reading of the particle rules is straightforward once we keep in mind the notion of usage
of the logical constant that they represent. A particle rule can be decrypted via these three
points163:
1. an X’s utterance,
2. a challenge, which is the demanding made by player Y over the initial X’s utterance,
3. a defense, corresponding to the answer of player X to the challenge made by Y.
Burden and/or object of choice
X-Utterance Y-Challenge X-defence de X
¬, there is no defence A |i : ¬ϕ A |i : ϕ ⊗
∧, the challenger has the
choice A |i : ϕ ∧ ψ
A |i : ?∧1
or A |i : ?∧2
A |i : ϕ respectively
A |i : ψ
∨, the defender has the
choice A |i : ϕ ∨ ψ A |i : ?∨
A |i : ϕ or
A |i : ψ →, both players shared the burden A |i : ψ → ϕ A |i : ψ A |i : ϕ
163 Except for the rule for negation since there is no defense.
157
Standard connectives – PR-SC
Particle Rules
Before presenting the structural rules, we have to deliver one more definition: that of
repetition rank. A repetition rank is a positive integer corresponding to the number of times
that a player can repeat the same challenge or the same defense164.
Structural Rules
Starting-Rule SR-0: Any play d∆ of a dialog D∆ starts with P uttering ∆ — the thesis. After the
utterance of the thesis, O has to choose a repetition rank. P chooses his repetition rank right
after O.
Playing-Rule SR-1: Players move alternatively. Each following the repetition rank is either a
challenge or a defense concerning a previous challenge.
Atomic Restriction SR-2: P cannot utter an atomic formula first. He is only allowed to re-use
those previously uttered by O.
Winning-Rule SR-3: A player X wins a play if and only if it is Y’s turn play but he cannot move
anymore with respect to the rules.
164 See Clerbout (2013, chap. 2) for a further discussion on repetition ranks.
158
Peut-on respecter une norme impossible?
JULIELE MARIA SIEVERS
Université de Lille (UdL 3)
UMR 8163 (Savoirs, Textes, Langage)165
Le rôle de la fiction dans le domaine normatif demeure un problème ouvert. Il en va de même
concernant la fonction exacte et la réelle nécessité de ce type de fiction. Sa présence, par
contre, est incontestable. On a, depuis le Droit Romain et jusqu’à nos jours, de nombreux
exemples d’usage des fictions dans la formulation des normes, dans les procédures et les
raisonnements juridiques :
“Substantive fictions became – and many continue to be – regular operations of the
law: In property law, a husband and wife could be treated as one person. In family law,
a child’s will is attributed to his guardian. In the interpretation of wills, one spouse may
be deemed to have predeceased the other, even though that may not in fact have
been the case. Under the attractive nuisance doctrine, a child who trespasses is
treated as having been invited onto the defendant’s land. In immigration law, an alien
may be considered to be legally excluded from the United States even though he is
physically within its borders. In civil forfeiture proceedings, property itself may be
named as a party to litigation. And most famously, a corporation may be treated as a
person, having the ability to sue and be sued, to buy and sell property, and to commit
torts; the Supreme Court has even recognized the corporation as a person under the
Fourteenth Amendment.” 166
165 Cf. Sievers, J. (2011). 166 “Notes” (2002, p. 2233-2234).
159
Ce type de fiction est habituel dans la tradition juridique, c'est-à-dire qu’on ne considère pas
très souvent le caractère ontologique fictif des normes, mais plutôt la présence des fictions
dans leur formulation. En effet, ces normes sont formulées via des termes qui n’ont aucune
référence dans le monde « réel », comme par exemple une personnalité qui ne concerne pas
une personne, mais une entreprise. Malgré le fait que ces formulations soient problématiques,
elles ne seront pas l’objet de notre présente étude167.
Nous allons plutôt nous concentrer sur des cas de « normes » qui revêtiraient en elles-mêmes
un caractère ontologique fictif, c’est-à-dire des cas où le caractère fictionnel ferait parti de la
notion normative elle-même. Un exemple notoire se trouve dans la notion de Norme
Fondamentale, proposée par le juriste autrichien Hans Kelsen (1881-1973). Cette ‘norme
fondamentale’ est selon lui une fiction indispensable, non seulement pour la compréhension
d’un ordre normatif légitime, mais essentiellement pour la justification du droit positif.
L’objectif de cette étude est d’approcher la notion de norme fictive par le biais de la notion
d’efficacité normative. Une norme, pour rester valide dans un ordre juridique, doit être
minimalement efficace, c'est-à-dire – observée par ses destinataires ou appliquée par le Droit.
Or, une norme fictive ne peut jamais être observée dans la réalité, car la fiction ne peut pas
être réalisée dans le monde. Ainsi, la norme fondamentale ne peut pas être à la fois une norme
et une fiction. Dans ce texte, nous défendrons l’idée qu’elle correspond uniquement au dernier
point, c'est-à-dire que la norme fondamentale est une fiction ; nous montrerons ensuite
pourquoi il n’est pas possible de soutenir le premier point, c'est-à-dire que nous ne pouvons
pas comprendre la norme fondamentale comme une norme juridique.
1. La norme fondamentale : sa fonction et ses caractéristiques
La définition de la notion de norme fondamentale a connu plusieurs phases dans les écrits de
Kelsen168. Dans cette étude, on analysera également les causes de cette évolution. Toutefois,
bien qu’évolutive, la définition de norme fondamentale de Kelsen est toujours donnée par
rapport à sa fonction dans le système normatif.
167 Cependant, ce problème sera traité brièvement dans la section 4 de ce texte. 168 Pour une brève description des tentatives de périodisation des approches de Kelsen, voir
Herrera, 1997, p. 30-34 et aussi Paulson, 1999.
160
En effet, la fonction de la norme fondamentale chez Kelsen est de fonder ainsi que de justifier
l’ordre comme un système de normes positives valides. Donc, elle ne se dirige pas vers une
conduite déterminée, elle ne prescrit pas une sanction : son objet (son « destinataire ») est
l’ordre normatif lui-même, dans son ensemble.
Cependant, afin de bien comprendre la justification d’ordre normatif, il est essentiel d’opérer
une distinction entre une norme et une proposition. Cette distinction nous mènera enfin à la
notion centrale de validité normative comme étant l’existence d’une norme169, et non une
propriété de cette norme. Par exemple, une proposition sur la réalité du monde est dite vraie
quand elle est confirmée par le moyen de notre expérience. Mais la norme n’est pas un
énoncé de l’être, elle n’est pas une proposition sur la réalité, et ne peut pas être vérifiée dans
le monde de la même manière. Tandis que la proposition est issue d’un acte de pensée, la
norme est issue d’un acte de volonté. Par conséquent, la norme n’est ni vraie ni fausse, au
même titre que des énoncés ayant la propriété d’être vrais ou faux : la norme peut seulement
être valide. La notion de validité est donc essentiellement liée à l’existence même de la
norme : la norme positive est valide si elle existe dans le cadre d’un ordre normatif, au sens
où elle est présente dans un système juridique spécifique. Dire d’une norme qu’elle n’est pas
valide n’a pas de sens: si la norme n’est pas valide, elle n’est pas une norme tout court. Seules
les « normes valides » peuvent être considérées comme ayant des propriétés juridiques
comme, par exemple, l’efficacité.
Un autre point essentiel de la théorie de Kelsen concerne les relations de dépendance entre
les normes juridiques. La justification de l’existence (ou de la validité) d’une norme
individuelle est réalisée par le biais de l’existence d’une norme plus générale dans l’ordre.
C'est-à-dire que le juge est autorisé à créer une norme individuelle au tribunal si elle est
soutenue par l’existence d’une norme plus générale de contenu correspondant. La validité de
cette norme générale est justifiée par l’existence d’une norme encore plus générale, et cette
régression continue jusqu'à la constatation, par exemple, de la Constitution d’un Etat. En fait,
c’est une erreur commune de confondre la norme fondamentale avec la première
Constitution historique d’un Etat, parce que la validité de la Constitution elle-même dépend
169 Il peut paraître insolite de défendre la thèse selon laquelle les normes existent, comme on le fait dans cette étude. On veut donc dès le début exprimer clairement notre façon de procéder : quand on parle de normes qui existent, on se limite aux normes selon l’approche positiviste, c'est-à-dire où les normes, pour pouvoir être appelées normes, doivent toujours être formulées ou être présentes dans un ordre juridique. C'est-à-dire que, pour le positiviste, une norme est une norme toujours par rapport à un système de normes qui est fondé sur une norme fondamentale. Ainsi, il n’est pas de notre intérêt d’investiguer ce qu’est la norme, ou les normes, ou de faire des réflexions métaphysiques autour de ce concept. On s’intéresse au problème de la possibilité d’existence des normes fictives (comme la norme fondamentale) dans le cadre d’un système positif, vu que la fictionnalité ne semble pas être un élément légitime pour le positivisme juridique.
161
encore d’une norme "plus haute", qui rend possible sa conformité au système comme un tout.
C’est ici que s’insère la notion de norme fondamentale. Grâce à la présupposition de la norme
fondamentale, le contenu de la Constitution peut être considéré comme valide et obligatoire,
vu que la Constitution est déjà un document concret élaboré par les hommes, et celle-ci n’est
pas valide en elle même. Elle est valide conformément à la norme fondamentale, qui est
présupposée être valide en tant que condition d’existence des autres normes de l’ordre
juridique. La fonction de la norme fondamentale est donc de donner les conditions
d’attribution des pouvoirs de création du droit, ainsi que son interprétation par la science du
droit. Kelsen dit :
La norme fondamentale n’est pas créée dans une procédure juridique par un organe
créateur de Droit. Elle n’est pas – comme la norme juridique positive – valide pour
avoir été créée de telle et telle façon par un acte juridique, mais elle est valide pour
être présupposée comme valide ; et elle est présupposée comme valide parce que
sans cette présupposition aucun acte humain ne pourrait être interprété comme un
acte juridique et, spécialement, comme un acte créateur de Droit.170
On voit clairement par ce passage le caractère essentiellement pragmatique de la norme
fondamentale : elle doit être présupposée parce qu’il n’y a pas d’autre manière de justifier
l’ordre comme un ordre juridique légitime. Sans la norme fondamentale, la science du Droit
ne pourrait ni analyser ni interpréter l’ordre juridique comme tel. C’est donc dans un domaine
scientifique que se trouve la réelle raison d’être de cette « norme ».
Ainsi, la probable formulation de la norme fondamentale est de cette façon liée à la
Constitution dont la validité est légitimée : les différents ordres normatifs possèdent
différentes normes fondamentales, et chacune de ces normes légitiment une Constitution
différente. De cette façon, dans le cas où une Constitution est supprimée – par une révolution
ou un coup d’Etat, par exemple – la norme fondamentale est automatiquement changée,
parce qu’elle se dirige vers une nouvelle Constitution, qui se substitue à l’ancienne. Avec une
nouvelle Constitution, le fondement de validité de tout cet ordre normatif doit, selon Kelsen,
aussi changer.
170 Kelsen (2005, p. 170). Traduit par l’auteur à partir de la version en Portugais.
162
2. Le début du problème : comment respecter une norme qui est une fiction ?
En relation avec la théorie artéfactuelle, proposé par Amie Thomasson (1999), on comprend
que les normes ont quelques caractéristiques en commun avec les objets fictifs (comme, par
exemple, les personnages d’un roman policier) : les normes positives sont elles aussi comme
des "abstract artifacts dependent on literary works and created by authors" (p. 148). Les
normes sont des objets abstraits dépendants du corps de loi, et elles sont créées par une
personne autorisée à le faire. Mais cela ne signifie pas encore que les normes sont "la même
chose" que les objets fictifs, c'est-à-dire qu’elles sont aussi des fictions. Les normes juridiques
ont quelques caractéristiques complémentaires qui empêchent cette dernière identification.
Les normes juridiques fondent leur existence d’une façon toute particulière. Elles doivent être
le sens d’un acte de pensée (comme les personnages littéraires, les concepts de la science…),
mais elles doivent être aussi le sens d’un acte de volonté (comme les impératifs, les
commandements). De plus, cet acte doit être objectif, c'est-à-dire qu’il doit être réalisé par un
sujet neutre, un tiers autorisé à créer la norme, à la rendre ainsi obligatoire pour ses
destinataires. Le juge crée les normes individuelles au tribunal, le législateur crée les normes
générales du "corps de loi"171.
En effet, les objets fictifs sont issus d’actes de pensée, ou "mental states". La norme
fondamentale est aussi le sens d’un acte de pensée "réel" et d’un acte de volonté fictif. C’est
pourquoi Kelsen s’autorise à dire qu’elle a des propriétés contradictoires qu’on analysera
ensuite. Or, il dit cela dans la phase de ses écrits où son objectif majeur est de bien différencier
les normes des propositions172! Cette affirmation de la part de Kelsen ne serait donc ni
gratuite, ni naïve. Celui-ci sait bien qu’il ne peut pas dire d’une norme qu’elle est
contradictoire, sans compromettre toutes ses conceptions sur le droit positif. Et pourtant, il
doit encore nommer "norme" la norme fondamentale, parce que les normes juridiques de
l’ordre ne peuvent être justifiées que par d’autres normes. Mais concernant les normes
juridiques standards, créées par un acte de volonté objectif, on ne peut rien dire de
contradictoire.
2.1 Piste pour la solution : l’efficacité d’une norme l’éloigne de la fictionnalité
171 Pour une excellente analyse de l’idée de production normative comme l’exercice d’un pouvoir chez Kelsen, voir Bobbio, 2008. 172 Voir Kelsen, 1996.
163
En effet, les normes expriment une prescription : quelque chose doit se produire dans le
monde173, et si cette chose ne se passe pas, une sanction doit être appliquée. Il faut donc
distinguer le cas où une chose doit se produire dans le monde du cas où une chose va se
produire. Ces deux champs – celui du devoir-être et celui de l’être – sont irréductibles, et on
ne peut pas traiter les normes de la même manière qu’on traite les énoncés sur les faits du
monde174. Mais ces deux domaines entretiennent encore d’autres types de relations, selon
Kelsen. La norme doit toujours, d’une manière générale, avoir une application dans le monde
pour rester valide. Cela veut dire que la norme doit être efficace, qu’elle doit être observée
par ses destinataires, ou encore que la sanction qu’elle prescrit doit être appliquée par les
organes prévus. Ceci ne signifie pas que la norme peut être réduite à cette conduite factuelle,
ou en être déduite. Quand une norme est appliquée ou observée, les deux domaines
expriment le même contenu concernant une conduite déterminée. Ce sont les contenus qui
sont en correspondance, et non les modes d’expression175 de la norme.
C’est donc dans un sens "pratique" que la norme a des rapports avec le monde. Cette
affirmation ne permet pas de dire qu’on pourrait traiter logiquement les normes de la même
manière qu’on traite les énoncés. Kelsen veut simplement dire qu’il est inutile et même
irrationnel de créer une norme qui ne puisse être observée de quiconque. Les référents de
ces normes dans le monde actuel sont ainsi les conduites qui ont été ou qui seront conformes
à ces mêmes normes ; c'est-à-dire que le contenu de la norme trouve ainsi une
correspondance dans le monde. De cette manière, l’application d’une loi révèle en soi sa
"vérification" dans le monde actuel, de sorte que les normes autorisant un sujet à effectuer
une opération interne du Droit sont constatées lors de la réalisation de cette opération.
Clairement, cela n’est pas la même chose de dire qu’un objet fictif « par excellence » comme,
par exemple, le personnage Leopold Bloom, est vérifié dans le monde (notamment dans les
rues de Dublin) au moment des célébrations du Bloomsday en hommage à James Joyce, tous
les 16 juin.
Bref, on peut dire que toutes les normes sont en effet limitées par la modalité du "possible",
l’important étant que le "possible" du Droit positif soit toujours actualisé à un certain
moment. Un bon exemple concerne le cas de la "loi sèche", qui prohibait la fabrication, le
173 La norme est plus qu’une commande ou un ordre, selon Kelsen : elle a aussi un caractère obligatoire parce qu’elle est produite par le juge, le sujet autorisé par le Droit. De cette façon, seul le Droit produit le Droit, et aucun élément externe n’est admis (autoproduction normative). Pour une discussion plus complète, voir Kelsen, 2005. 174 Sur le sophisme naturaliste, voir Hume, 2000. 175 Voir la discussion sur le substrat indifférent au mode, voir Kelsen, 1996. Pour une autre approche sur cette même question, voir Hare, 1996.
164
commerce, le transport, l’importation et l’exportation de boissons alcoolisées aux Etas Unis
des années 20. Au Brésil, et dans d’autres pays, une loi similaire est destinée aux conducteurs
de voitures. Pourtant, le manque d’inspection de la part des policiers dans certaines régions
a eu des répercussions dans le respect de la norme par les citoyens, et on peut dire
qu’aujourd’hui que la norme a presque perdu sa validité dans certaines régions, à cause de la
perte de son efficacité. Cependant, la "loi sèche" des Etas Unis a complètement perdu sa
validité en 1933 par son abrogation, également en raison de son manque d’efficace176. On
constate ainsi une profonde interaction entre le domaine normatif et le monde actuel : le
monde actuel est changé à partir des normes, mas le domaine normatif est aussi changé à
partir des faits. Toutefois, cela ne concerne pas les objets fictifs, vu que la fiction est
essentiellement détachée de la réalité du monde actuel. Même si une histoire peut traiter
avec beaucoup de vraisemblance d’un événement réel, cette histoire n’est pas destinée à
changer quelque chose dans la réalité177. Cette dynamique concernant la relation constante
entre les faits et les normes est déjà une caractéristique qui empêche l’insertion de l’élément
de la fiction dans le Droit. Sans l’actualisation de ce qui est possible selon le devoir-être dans
le monde actuel, la norme perd son efficacité. Elle perd alors également sa validité, c'est-à-
dire qu’elle n’existe plus. La notion de validité normative, comme ayant besoin d’un minimum
d’efficacité dans le monde actuel, rejette donc par elle-même le caractère fictif.
Ainsi, une piste pour la réponse se trouve dans le fait que les normes fictives ne peuvent pas
être observées par ses destinataires. La norme a besoin d’être efficace pour continuer à
exister comme une norme valide. Cela pose clairement des obstacles à la présence de la fiction
dans le domaine normatif. Si la norme fondamentale est une fiction par excellence selon
Kelsen, on a des raisons de suspecter son statut de norme.
2.2 La réponse : les normes fictives ne sont pas des normes. On ne peut donc pas les
observer, les violer, ou les appliquer.
176On ne défend pas que toutes les normes d’un ordre sont efficaces. Dans nombres d’ordres demeurent des normes qui ne sont jamais appliquées. Mais on est ici dans la théorie du Droit de Kelsen, qui stipule comment le Droit doit être. Décrire comment le Droit effectivement procède relève de la sociologie du Droit. 177 Certes, on peut trouver des exemples d’œuvres littéraires qui semblent avoir un caractère fortement normatif. Mais on soutient que, en général, ce n’est pas le but spécifique de la littérature, c'est-à-dire que cela n’est pas ce que normalement on attend d’une œuvre littéraire.
165
Kelsen n’hésite pas à reconnaître quelques problèmes (mais qui seraient simplement
apparents, selon lui) liés à la fonction et l’importance de la norme fondamentale. Un des
problèmes consiste dans le fait que, en tant que point culminant de la recherche pour la
justification juridique, la norme fondamentale ne peut pas avoir le même caractère que toutes
les autres normes de cet ordre. Si la norme était formulée de manière standard, elle aurait
aussi besoin d’être elle-même justifiée. Donc, la norme fondamentale gagne un statut
différencié ; elle est placée à un niveau "panoramique" par rapport à l’ensemble de l’ordre.
Pour justifier l’ensemble des normes, en donnant l’unité du système, la norme fondamentale
est seulement présupposée. Kelsen nous répond qu’elle doit être le sens d’un acte de volonté
fictif. Cela veut dire que cette norme est conçue comme étant consciemment créée par le
sujet compétent comme une norme fictive, avec un propos particulier bien défini, à savoir
avec l’intention de la légitimation du système juridique. Cette norme est impossible parce que
sa propre production indique une erreur au moment de la production d’une norme : la norme
fondamentale ne reçoit pas la légitimation d’une norme plus "haute", le juge n’est pas
autorisé à produire une telle norme. Dans ce sens elle doit être considérée comme une fiction
selon Kelsen, parce qu’il est impossible d’observer ou d’appliquer une norme qui ne fait pas
partie d’un système, et qui est seulement supposée par son auteur avec un propos
méthodologique extérieur aux démarches internes du droit. Le juriste n’applique pas la norme
fondamentale, il fait appel au recours méthodologique de la norme fondamentale quand il en
a besoin, il utilise la notion de norme fondamentale. Ainsi, la norme fondamentale n’est pas
une norme à observer, mais seulement un apparatus cognitif qu’on regarde comme une
norme (fictive) pour pouvoir considérer l’ordre normatif comme un ensemble cohérent et
possible à interpréter. Ainsi, le rôle principal joué par la norme fondamentale est de légitimer
la possibilité de la création normative. C’est-à-dire légitimer le fait de pouvoir arriver à une
norme fondamentale dans la recherche pour la justification normative, ce qui atteste que
toutes les conditions pour la création normative ont été remplies.
Par contre, on ne peut pas dire que les normes juridiques « ordinaires » soient fictives dans
ce sens là. Certes, elles n’ont pas d’existence matérielle ou physique dans le sens où l’on
pourrait voir une norme dans la nature, mais elles existent de la même manière que certains
objets abstraits existent. Bruno Celano178 écrit : “…according to Kelsen, law is norm, and norms
are the content (or, the sense: Sinn) of intentional acts, or states, directed at somebody else’s
behavior. A legal order is a complex, structured, sense content. Its mode of existence may be
compared to the mode of existence of a work of art, or a scientific theory.” Ainsi, le même type
178 Celano (2000, p. 174).
166
de problème semble concerner d’autres “objets” qui, même si n’étant pas présents
visiblement dans la réalité du monde, ne peuvent pas non plus être considérés comme étant
des fictions, comme les théories scientifiques, par exemple. Il serait vraiment étrange de dire
qu’une hypothèse scientifique devienne, après son acceptation par le consensus au sein d’une
communauté scientifique, une fiction.
Une norme juridique spécifique possède, selon Kelsen, une validité (qui est, comme on a déjà
cité, son existence spécifique) temporelle et spatiale (territoriale). On peut dire, par exemple,
qu’en Angleterre il existe une norme qui permet aux couples de même sexe d’adopter des
enfants ; cette norme est donc présente dans l’ordre juridique de ce pays, elle est valide là-
bas. En Afrique, par exemple, cette norme n’existe pas, et elle n’a jamais existé. On peut
également dire qu’il y avait dans plusieurs pays, selon différents Constitutions, des normes
légitimant l’esclavage, et qu’aujourd’hui ces normes n’existent plus nulle part.
La norme « commence » à exister dans un moment donné, et elle peut cesser d’exister dans
un autre moment donné, par abrogation par exemple. Et cette existence normative est même
vérifiable, dans le même sens qu’on peut vérifier l’existence d’un élément abstrait, comme
l’état fébrile d’un enfant, par exemple, qu’on vérifie en touchant son front ou en regardant le
thermomètre. On vérifie l’existence d’une norme positive en regardant les documents
juridiques d’un Etat, ou le règlement d’un club, ou certains versets de la bible, par exemple.
Ces normes ont une validité spatiale (même si cette validité est indéterminée, p. ex. partout)
et temporelle (même si elle est indéterminée, p. ex. pour toujours). Cette notion particulière
d’existence ne place pas la norme dans le domaine naturel de l’être. On dit seulement qu’il
existe une norme quelconque qui prescrit (oblige, interdit…) une conduite, mais on ne dépasse
pas l’abîme entre l’être et le devoir-être, fortement estimé par Kelsen. Ainsi, on est seulement
autorisé à dire, par exemple, que « la proposition ‘Il existe une norme selon laquelle tous les
citoyens doivent payer leurs impôts’ est vraie ». Kelsen179 dit :
Affirmer qu’une norme est valide, c’est toujours affirmer qu’elle est valide pour un certain
espace et un certain temps ; c’est-à-dire qu’elle se rapporte à des événements qui peuvent se
dérouler seulement quelque part et à un moment donné. La relation d’une norme à l’espace
et au temps constitue le domaine de validité spatial et temporel de cette norme ; ce domaine
de validité peut être soit limité, soit illimité. (…) un exemple typique de domaine de validité
restreint est le domaine de validité des normes d’un ordre juridique étatique, qui est présenté
comme un élément de l’Etat, en tant que « territoire de l’Etat ».
179 Kelsen (1996, p.191).
167
On verra ensuite comment Kelsen a caractérisé la norme fondamentale dans deux phases de
ses écrits. Malgré la reformulation de la définition de cette norme par l’auteur, un aspect
demeure inchangé : la norme fondamentale ne fait pas partie du système juridique avec les
autres normes. Elle possède des caractéristiques assez spécifiques pour ne pas pouvoir être
classée parmi elles. Ces caractéristiques la relient surtout à la science du Droit, et non avec
son objet, les normes juridiques.
3. Hypothèses versus Fictions
On arrive maintenant à la considération d’une approche centrale pour Kelsen à propos de la
fiction, à savoir, la philosophie du "comme si" de Hans Vaihinger. Cette approche sert de base
à la conception de norme fondamentale comme une fiction de la Science du Droit, et elle a
influé sur la formulation kelsénienne de la notion de norme fondamentale comme ayant des
contradictions. Pourtant, on analysera d’abord la première conception kelsénienne de norme
fondamentale comme hypothèse, sous l’influence d’Emmanuel Kant.
3.1 La norme fondamentale comme hypothèse : l’influence de Kant
Pour Kelsen, selon son œuvre posthume Théorie Générale des Normes (1979), la norme
fondamentale n’est pas une norme positivée, elle n’est pas établie par un individu, car
personne n’est autorisé à la produire de cette façon (il n’y a pas d’autre norme plus haute qui
autorise quiconque à la produire). Elle est une norme présupposée dans la pensée juridique.
Mais la norme fondamentale n’est pas seulement le sens d’un acte de pensée : elle est aussi
le sens d’un acte de volonté : un acte de volonté fictif.
Mais avant cela, dans la Théorie Pure du Droit (1962, 2ème éd.), Kelsen révèle une certaine
influence de Kant, quand il fait une comparaison entre la justification de la norme
fondamentale et la justification des concepts purs de l’entendement. Dans le long passage ci-
dessous, on reconnaîtra la première position de Kelsen sur les caractéristiques de la norme
fondamentale, qui la relient à la philosophie transcendantale kantienne, en assumant la
168
norme fondamentale comme hypothèse logico-transcendantale pour l’interprétation des
normes de l’ordre par la science du Droit :
En tant que l’hypothèse d’une certaine norme fondamentale permet seule de
reconnaître à l’acte ou au fait constituants, et aux actes ou faits réalisés
conformément à cette Constitution, la signification objective qui concorde avec leur
signification subjective, c’est-à-dire de leur reconnaître le caractère de normes
juridiques objectivement valables, on peut dire que la norme fondamentale que
présente la science du droit180 est - s’il est permis d’utiliser par analogie un concept
de la théorie kantienne de la connaissance – la condition logique transcendantale de
cette interprétation. De même que Kant pose la question de savoir comment une
interprétation des faits donnés à nos sens exempte de toute métaphysique est
possible dans les lois naturelles formulées par les sciences de la nature, de même la
théorie pure du droit pose la question de savoir comment est possible une
interprétation de la signification de certains faits comme une système de normes
juridiques objectivement valables, descriptibles dans des propositions de droit, une
interprétation qui ne recoure pas à des autorités métajuridiques telles que Dieu ou la
Nature.
A cette question de théorie de la connaissance, la théorie pure du Droit répond : une
telle interprétation est possible à la condition de supposer la norme fondamentale,
« on doit se conduire comme la Constitution le prévoit », c’est-à-dire de la façon qui
correspond au sens subjectif de l’acte de volonté constituant, aux prescriptions ou
commandements du législateur constituant. 181
La norme fondamentale du Droit, en tant que supposition, peut être comprise comme une
hypothèse. Néanmoins, un détail interdit cette façon de la comprendre : cette norme garde
dans sa propre formulation des éléments qui contredisent la théorie dont elle est issue. La
norme fondamentale n’est pas une simple supposition, elle est aussi un type de formulation
en contradiction avec les conditions de sa propre production. Sa "production" est une
infraction à la production normative. Autrement dit, la norme fondamentale n’est pas une
hypothèse parce qu’elle n’est même pas possible dans l’ordre juridique.
180 On marque cette partie du passage pour accentuer le fait que Kelsen lui-même situe la norme fondamentale dans le contexte de la science, et non dans le Droit. 181 Kelsen (1962, p.201-202).
169
3.2 La norme fondamentale comme fiction : l’influence de Vaihinger
Dans l’ouvrage Théorie Générale des Normes, Kelsen ira jusqu’à abandonner et même nier
l’identification de la norme fondamentale avec une hypothèse. Dans cette nouvelle phase, il
revisite la notion de norme fondamentale sous l’éclairage de la philosophie du "comme si" de
Hans Vaihinger :
Comme il ressort de ce qui précède, la norme fondamentale d’un ordre moral ou
juridique positif n’est pas une norme positive, mais une norme simplement pensée,
c’est-à-dire une norme fictive, la signification d’un acte de volonté non pas réel, mais
fictif. En tant que telle, elle est une fiction véritable au sens de la « philosophie du
comme si » de Vaihinger, une fiction caractérisée par le fait que non seulement elle
contredit la réalité, mais aussi qu’elle est contradictoire en soi. En effet, l’hypothèse
d’une norme fondamentale – par exemple la norme fondamentale d’un ordre moral
religieux : « On doit obéir aux commandements de Dieu » ou la norme fondamentale
d’un ordre juridique : « On doit se comporter comme le détermine la Constitution
historiquement première » - contredit non seulement la réalité, car aucune norme de
ce genre n’existe en tant que signification d’un acte de volonté réel, mais aussi elle
est contradictoire en soi, car elle représente l’habilitation d’une autorité juridique ou
morale suprême, et part ainsi de l’idée d’une autorité (toutefois purement fictive) plus
haute encore que cette autorité. Une fiction est, selon Vaihinger, un expédient de la
pensée dont on se sert si l’on ne peut pas atteindre le but de la pensée avec les donnés
dont on dispose. Le but de la pensée de la norme fondamentale est le fondement de
la validité des normes constituant un ordre moral ou juridique positif, c’est-à-dire
l’interprétation de la signification subjective des actes posant ces normes comme leur
signification objective (c’est-à-dire comme normes valides) et les actes concernés
comme des actes posant de normes. Ce but ne peut être atteint que par la voie d’une
fiction. C’est pourquoi il convient de remarquer que la norme fondamentale, au sens
de la « philosophie du comme si » de Vaihinger n’est pas une hypothèse – comme je
l’ai moi-même quelquefois caractérisée –, mais une fiction qui se distingue de
l’hypothèse par le fait qu’elle est accompagnée ou doit être accompagnée de la
conscience que la réalité ne lui est pas conforme. 182
182 Kelsen (1996, p.344-345).
170
Quand les critiques parlent de la norme fondamentale kelsénienne comme d’une fiction, ils
ont rarement à l’esprit la notion de fiction de la manière dont Kelsen l’a caractérisée, c'est-à-
dire une fiction au sens de la philosophie de Vaihinger. De cette manière, il est commun
d’entendre parler de la norme fondamentale comme une fiction, la considérant implicitement
comme un objet non-concret. Cependant, on est bien d’accord sur le fait que la norme
fondamentale est un objet abstrait, comme le sont aussi toutes les autres normes. Mais cela
n’implique pas que ces normes soient aussi des fictions. Du moins, cela n’est pas du tout en
accord avec ce que Kelsen a proposé. Sa conception de fiction s’engage essentiellement avec
l’élément de la contradiction et, ainsi, la norme fondamentale est placée aux cotés des
énoncés d’être, et non au milieu des normes de Droit.
La norme fondamentale comme fiction selon l’interprétation de Kelsen nous contraint à
reconnaître l’élément de la contradiction en ce sens qu’elle n’existe pas et ne peut pas exister
dans la réalité de l’ordre juridique. C'est-à-dire que la norme est en contradiction avec la
réalité parce qu’elle n’est pas une norme formellement présente dans l’ordre, puisqu’elle n’a
pas pu être créée au moyen d’un acte de volonté réel, d’un acte subjectif dans le monde des
faits objectivé par la figure du législateur. De plus, Kelsen dit que la norme est "contradictoire
en soi", parce que la formulation même de la norme fait appel à une autorité plus haute que
celle du législateur. Cette autorité servirait à formuler la norme comme telle. Ainsi, la présence
des contradictions empêche l’utilisation du terme "hypothèse" pour caractériser la norme
fondamentale. En effet, une hypothèse, dans une théorie scientifique, ne peut pas contredire
la réalité. En revanche, elle doit partir des présuppositions constatables, pour rechercher un
résultat déterminé, qui viendra confirmer ou réfuter la présupposition. Même si une
hypothèse peut avoir une caractéristique apparemment fictionnelle (dans le sens qu’elle n’est
pas encore présente dans la réalité, mais seulement supposée être présente afin qu’on puisse
vérifier l’état du monde si l’hypothèse était soutenue), elle ne s’engage pas avec des
contradictions. Autrement dit, Kelsen doit éliminer la notion d’hypothèse parce que
l’élaboration de la norme fondamentale se donne avec la conscience même de ce que cette
"opération" n’est pas conforme aux procédures de la production normative. Ainsi, on constate
une espèce d’exception pendant la formulation de la norme comme fiction, on fait un
"attentat" aux principes positivistes de la production normative. Mais son but est justifié par
les nécessités théoriques de clôture et de légitimation de l’ordre juridique, attestées par la
science du Droit.
171
J. Loewenberg183 démontre de manière claire l’opposition entre la notion de fiction soutenue
par Vaihinger, face à la notion d’hypothèse, et face aussi à d’autres notions de fiction, comme
celles de la théorie artéfactuelle, par exemple :
Fictions, in Vaihinger’s usage, are not identical with figments, such as centaur or fairy,
nor are they hypotheses capable of verification. They are deliberate devices
(Kunstgriffe) on the part of thought for the practical purpose of successful orientation
in and perfect control over the environment. Theoretically they are absolutely
valueless. Applied with a knowledge of their fictitious character, they will lead to the
intended practical results, but used as hypotheses, they must necessarily create
confusion and false theories, for a fiction is defined as that which is both contradictory
in itself and which has no correspondence with reality (sensational flux).
Ce passage illustre l’aspect d’utilité pratique de la fiction selon Vaihinger, et cette
caractéristique est central pour l’identification de la fiction selon cet auteur.
4. Normes fictives et personnages fictifs
Pour Kelsen, comme on l’a déjà dit, la validité d’une norme réside dans le domaine du devoir-
être, et ne peut pas se réduire au domaine de l’être. Mais Kelsen n’admet pas que la validité
soit complètement détachée de la réalité : un ordre juridique valide doit être dans un
minimum efficace, c’est-à-dire que les normes doivent être un minimum respectée ou
observée. Une norme ne peut jamais être "fausse" : si la norme est valide, elle existe, elle est
présente dans l’ordre. La norme fondamentale n’a rien à voir avec la réalité (et même, elle a
une relation contradictoire avec la réalité), au contraire de certaines normes qui ont pour
objectif de régler la conduite humaine. Pour ces normes, un minimum d’efficace est exigé.
Contrairement à cela, Otto Pfersmann184 dit : « La loi peut être limitée par la constitution
comme le règlement peur être limité par la loi, mais ni la constitution, ni la loi, ni le règlement
ni aucun autre acte juridique ne sont limités, à première vue, par la réalité physique ». On ne
peut pas être d’accord avec cela car, pour pouvoir être efficace, c'est-à-dire pour qu’on puisse
observer, respecter ou appliquer la norme, on doit pouvoir réaliser dans le monde ce qui est
183 J. Loewenberg (1912, p.717). 184 Pfersmann (2004, p.43).
172
prescrit ou interdit par la norme. La norme fondamentale peut être à son tour impossible
parce qu’elle ne se dirige pas vers les personnes dont les conduites se passent dans le « monde
réel », mais elle se dirige vers un ordre normatif qui appartient au domaine du devoir-être.
On soutient une approche encore plus intégrale sur la notion de fiction et le droit, à savoir,
que les normes avec des fictions dans leur formulation (les exemples plus courants de fiction
juridique) ne sont pas de caractère fictif de la même manière que, par exemple, un
personnage littéraire est fictif. Ainsi, dans cet article, on pose les limites de la comparaison
entre le Droit et la Littérature de manière très nette et rigoureuse. Une norme qui présente
des fictions peut être simplement reformulée sans présenter ces fictions, sans aucune
altération de son sens. Or, on peut attester l’existence des normes dites fictives depuis
l’Antiquité, et surtout avec des exemples du Droit Romain185. Dans le cas de la fiction
normative qui dit, par exemple, qu’un fils adoptif doit être considéré comme un fils légitime,
celle-ci peut être exprimée, par exemple, en disant que le fils adoptif doit avoir les mêmes
droits et devoirs qu’un fils légitime. La fiction est, de cette façon, une simple manière
d’exprimer une chose pour une autre. Mais on ne peut pas dire la même chose concernant un
objet fictif comme le Minotaure ou Sherlock Holmes : Sherlock Holmes ne peut exister qu’en
tant que personnage fictif crée par Conan Doyle. Mais la norme dite fictive peut encore exister
comme norme, en s’adressant aux mêmes destinataires et avec le même sens, sous une autre
formulation normative. Certainement, la formulation normative sans le recours à la fiction
contraindrait à des changements courants et importants, rarement envisageables dans les
codes et les textes de loi. C’est précisément ce point qui légitimerait l’usage des fictions
juridiques. Or, le même Otto Pfersmann (2004) dénonce le caractère arbitraire de cet usage,
quand il nie l’existence des fictions juridiques pour défendre la thèse, condamné dans ce texte,
de la fictionnalité du Droit comme un tout. Pour renforcer notre critique à l’égard de la
formulation normative par le biais des fictions, on peut citer l’analyse de Pfersmann de la
fiction juridique selon laquelle "Seules les femmes ont accès à cette piscine, le maître-nageur
est une femme au sens de ce règlement":
Dans l’exemple de la piscine, il n’est pas dit qu’un homme est une femme, mais
simplement que, au cas où il existe un maître-nageur du sexe masculin, cette
personne doit être traitée comme une personne du sexe féminin au sens du règlement
de la piscine de N***, en d’autres termes, il lui est permis de pénétrer dans la piscine.
Le même résultat aurait été obtenu, si le règlement avait dit :
185 Voir aussi Guerrier, 2004.
173
N : Seules les femmes sont admises dans cette piscine, à l’exception du maître-nageur
en charge de la sécurité des personnes et des installations.
Or, il s’agit simplement d’une autre formulation de la même norme utilisant une autre
technique. 186
Concevoir la fiction comme une technique signifie la concevoir comme un outil pour le Droit,
de la même manière qu’elle est un outil pour la Physique ou la Chimique. Mais le problème
est que la norme fondamentale ne peut se présenter d’autre façon que par la fiction. Cela
nous contraint à poser cette « norme » en dehors du système normatif.
5. Les fictions ne sont jamais des normes, les normes ne sont jamais des fictions
Dans le contexte du positivisme de Kelsen, quand l’auteur dit que la norme fondamentale est
une fiction, il parle donc de la fiction dans le sens vaihingérien. La norme fondamentale, et
elle seule, possède des contradictions dans sa définition même, parce qu’elle ne consiste pas
en le sens d’un acte de volonté réel. Cette caractéristique nous permet de traiter la norme
avec des prédicats destinés aux propositions. Une norme positive ne peut pas avoir de
contradictions, parce qu’elle est seulement valide, et non vraie ou fausse. Autrement dit, la
norme fondamentale est surtout une construction théorique, une "construction auxiliaire de
la pensée", et se destine aux propos théoriques de la science du Droit.
En ce sens on doit comprendre la norme fondamentale comme une fiction pour le Droit, car
le Droit lui-même ne peut pas contenir des normes fictives, marquées par des contradictions,
selon les conceptions de Vaihinger et Kelsen. La norme fondamentale n’est pas dirigée vers la
réalité des conduites humaines, elle est dirigée vers l’intégrité de l’ordre normatif. Le contenu
d’une norme juridique doit être observé ou appliqué dans la réalité du monde pour que la
norme qui prescrit ce contenu continue à être une norme ; mais la norme fondamentale est
destinée à la validité de l’ordre comme un tout, et elle n’est pas "appliquée" de la même façon.
Une norme fondamentale n’a pas cette même relation avec le monde actuel, et sa validité ne
dépend pas de son efficacité. Sa validité est une fiction, et se trouve dans le champ de
l’impossible.
186Pfersmann (2004, p.54)
174
Conclusion
L’immense travail de Kelsen pour construire une théorie "pure" pour le droit (et même pour
le domaine normatif en général) a parfois rencontré des problèmes internes tout au long de
ses écrits. Kelsen lui-même n’a jamais hésité à les révéler et à corriger ses thèses. Il y a même
eu des moments où Kelsen a changé radicalement ses approches sur des problèmes pertinents
de la théorie du Droit. C’est à cause de cela que, concernant différentes notions ou positions
kelséniennes, on peut bien identifier des "phases" dans la pensée de cet auteur.
De même pour la notion de norme fondamentale. Cette notion, probablement la plus discutée
et attaquée par les critiques de Kelsen, est analysée et défendue ici comme une fiction pour
le droit. De plus, on considère aussi que dans le droit positif il n’y a pas de place pour les
fictions normatives.
Dans la présente analyse des caractéristiques particulières de la norme fondamentale, un des
aspects les plus importants est la différenciation entre les normes et les propositions qui les
décrivent. Cette différence nous a permis de reconnaître dans la norme fondamentale (en
accord avec la lecture kelsénienne de Vaihinger) quelques caractéristiques des propositions,
telles que la contradiction, vu que la norme fondamentale est le résultat d’un acte de pensée
« réel » et d’un acte de volonté fictif.
Dans cette perspective, on a alors considéré la notion de fiction sous l’approche de la théorie
artéfactuelle pour la comparer à des fictions littéraires, en montrant que les normes ordinaires
ne peuvent pas être comparées à ces fictions. Ensuite, on a analysé la notion de norme
fondamentale comme hypothèse. En effet, si on considère l’analyse de cette norme en tant
que simple présupposition du législateur, comme élément qui n’est pas présent dans l’ordre
juridique, on peut noter une similitude avec la notion d’hypothèse. Toutefois, cette
identification est en fait impropre, en raison de l’existence d’éléments contradictoires qui
modèlent la notion de norme fondamentale : elle possède des caractéristiques contradictoires
avec sa propre formulation. Le constat de cette erreur a conduit Kelsen à la théorie
vaihingérienne du "comme si", et à la conception finale de norme fondamentale comme
fiction.
On a constaté que l’approche de Kelsen de la norme fondamentale comme porteuse de
contradictions, une fois insérée dans une science avec un propos pratique bien défini, est
"autorisée" par la théorie de Vaihinger. C'est-à-dire que selon cette théorie, seule la norme
175
fondamentale peut être considérée comme une fiction légitime, et non pas les autres normes
de l’ordre.
Finalement, on conclut que la notion de norme fondamentale comme fiction est appropriée
aux propos de la théorie kelsénienne pour le droit, et qu’elle atteint son objectif de
justification du système juridique en tant que "séparée" du droit, ayant le propre droit comme
objet. On soutient en outre qu’aucune norme de l’ordre positif ne peut être une fiction : si
l’on comprend la fiction comme un manque de référence (Thomasson), on s’appuie sur la
nécessité de l’efficacité normative.
Si on comprend cette notion comme un outil méthodologique qui présente des
caractéristiques contradictoires pour expliquer par supposition ce qui ne peut pas être
expliqué dans la réalité (Vaihinger), on s’appuie sur le fait que les normes juridiques ne
peuvent pas avoir de contradictions. La constatation de la présence des contradictions dans
la formulation de la norme fondamentale, parfois considérée problématique, est en fait
précise. Cette présence ne viole pas les processus scientifiques qui sont réalisés par cette
« norme ». Par conséquent, les contradictions présentes dans la formulation de la norme en
tant que fiction permettent de légitimer les démarches juridiques au niveau de la science du
droit, pour la justification de la validité de l’ordre normatif dans son ensemble. Autrement dit,
c’est essentiellement la formulation de la norme fondamentale comme une fiction qui rend
possible l’existence légitimée d’un système juridique positif.
Références bibliographiques:
Atienza M. [ 2003]: "As razões do Direito. Teorias da Argumentação Jurídica".
Tradução de Maria Cristina Guimarães Cupertino. São Paulo: Landy Livraria e Distribuidora
Ltda.
Barzotto L. F. [1999]: "O Positivismo Jurídico Contemporâneo. Uma introdução
a Kelsen, Ross e Hart". São Leopoldo: UNISINOS.
Bobbio N. [2008]: "Direito e Poder". Tradução Nilson Moulin. São Paulo: Editora UNESP.
Bobbio N. [1997]: "Teoría General del Derecho". Santa Fé de Bogotá –
176
Colômbia: Editorial Temis.
Cabrera C.A. [2000]: "Lecciones de lógica jurídica". Espanha: Editorial Mad, S.L.
Celano B. [2000]: "Kelsen's Concept of the Authority of Law", In: Law and Philosophy, Vol. 19,
No. 2, Kelsen, Authority and Competence (Mar., 2000), pp. 173-199
Goyard-Fabre S. [1993]: "Kelsen e Kant. Saggi sulla dottrina pura del diritto".
Napoli: Edizioni Scientifiche Italiane.
Guerrier O. [2004] : "Fictions du Droit et Espace Littéraire". In Fictions du
savoir à la Renaissance, URL : http://www.fabula.org/colloques/document88.php
Hare R. W. [1996]: "A Linguagem da Moral". Tradução de Eduardo Pereira e
Ferreira. São Paulo: Martins Fontes.
Herrera C. M. [1997] : "Théorie Juridique et Politique chez Hans Kelsen".
Editions Kimé, Paris.
Hume D. [2000]: "Tratado da Natureza Humana". São Paulo: Fundação Editora
da UNESP (FEU).
Kaufmann E. [1992]: "Critica della filosofia neokantiana del diritto". Napoli:
Edizioni Scientifiche Italiane.
Kelsen H. [2005]: "Teoria Geral do Direito e do Estado". São Paulo: Martins
Fontes, [1° ed. 1945].
______. [1962] : "Théorie Pure du Droit". Paris : Dalloz, [1° ed. 1960].
______. [1996] : "Théorie Générale des Normes". Paris : Léviathan, [1° ed.
1979].
Loewenberg J. [1912]: Reviewed work(s): "Die Philosophie des Als Ob. System der
Theoretischen, Praktischen und Religiösen Fiktionen der Menschheit auf Grund Eines
Idealistischen Positivismus. Mit einem Anhang über Kant und Nietzsche" by H. Vaihinger. The
Journal of Philosophy, Psychology and Scientific Methods, Vol. 9, No. 26 (Dec. 19, 1912).
Published by: Journal of Philosophy, Inc. Stable URL: http://www.jstor.org/stable/2013048
"Notes" [2002]: "Lessons from Abroad: Mathematical, Poetic, and Literary Fictions in the Law".
In Harvard Law Review, Vol. 115, No. 8, Jun., 2002, p. 2233-2234.
Paulson S. [1999]: "Arriving at a Defensible Periodization of Hans Kelsen's
Legal Theory". In: Oxford Journal of Legal Studies, Vol. 19, No. 2 (Summer, 1999), pp. 351-364.
Pfersmann O. [2004] : "Les modes de la fiction : Droit er Littérature". In : Lavocat, F. Usages et
théories de la fiction, PUR.
177
Ross Alf. [2003]: "Direito e Justiça". Tradução Edson Bini. Bauru, SP: EDIPRO.
Thomasson A. [1999]: "Fiction and Metaphysics". Cambridge University Press,
Cambridge.
178
Kelsen contre les fictions dans le Droit
Juliele Maria Sievers
Université de Lille 3
UMR 8163 (Savoirs, Textes, Langage)
La notion de fiction est normalement considérée comme un élément légitime non seulement
dans la Littérature, mais aussi dans le domaine normatif, surtout le Droit, depuis les exemples
bien connus des fictio legis du Droit Roman. Pourtant, une analyse concernant le statut
ontologique des normes dites fictives révèle quelques aspects que nous permettent dire que
le domaine normatif n’est pas un contexte légitime pour la fiction. Dans ce texte, on analysera
la notion de fiction dans le contexte du positivisme juridique de Hans Kelsen, avec l’objectif
de démontrer que aucune des normes juridiques positives ne peut pas être appelée « fictive ».
On se servira de la notion de norme fondamentale, une fiction selon Kelsen, pour démontrer
que la fiction doit être vue, dans ce contexte, comme un outil méthodologique pour la
compréhension de l’objet d’une science, et on démontrera aussi qu’aucune approche sur la
notion de fiction ne peut être appliquée directement aux normes. Ainsi, on utilise la notion de
norme fondamentale comme une fiction pour le Droit, pour attaquer la notion de normes
fictives. Cette approche éclairera aussi la propre interprétation de la notion de norme
fondamentale comme une fiction selon Kelsen, puisque cette notion est la cible de plusieurs
critiques, certaines desquelles on essayera de répondre dans ce texte. On conclura que la
fiction, en tant que technique juridique, ne peut pas être au même temps un objet du Droit,
et que les normes dites « fictives » ont d’ailleurs le même statut ontologique que toutes les
autres normes juridiques.
179
Introduction
L’objectif de ce texte c’est d’analyser la notion de norme fondamentale comme une fiction
dans la théorie de Hans Kelsen. Plus précisément, l’objectif est de démontrer que le
positivisme juridique de Kelsen187 ne permet pas l’existence de normes fictives, et que la
norme fondamentale ne fait pas partie du Droit, c'est-à-dire ici, du corps normatif d’un Etat.
Après présenter de manière générale la notion de norme fondamentale, avec les
caractéristiques données par Kelsen, et que sont conservés à travers tous ses écrits, on
essayera d’analyser le statut ontologique de la norme fondamentale sous l’approche de la
théorie artéfactuelle de la fiction (Thomasson, 1999), pour montrer que la norme
fondamentale est bien d’accord avec la notion de fiction soutenue par cette auteure,
différemment des normes du système juridique. L’approche de la théorie artéfactuelle nous
servira de base pour la défense d’une conception anti-fictionnelle pour le Droit. On analysera
aussi les principaux problèmes avec la qualification initiale kelsénienne de cette norme
comme une hypothèse, conception souligné par l’influence d’Imannuel Kant. Ensuite, on
analysera les motivations qui ont conduit Kelsen à la conception finale de cette norme comme
une fiction, sous l’influence de la philosophie de Hans Vaihinger, pour démontrer que la fiction
utilisé par Kelsen ne s’insère pas dans le domaine normatif. Cette dernière approche de Kelsen
est actuellement la cible de certaines critiques, lesquelles on essayera de répondre dans ce
texte par le moyen de notre thèse « anti-fictionnelle » pour le Droit. Enfin, l’enjeu principal de
ce texte est de présenter la thèse de que la notion de norme fictive n’est pas légitime dans le
Droit, et on soutiendra cette thèse par le moyen de trois principaux arguments.
Premièrement, une analyse du statut ontologique des normes et de la norme fondamentale
selon une approche contemporaine, à savoir, la théorie artéfactuelle, nous montrera que la
norme fondamentale a les mêmes caractéristiques ontologiques que les objets fictifs
littéraires, par exemple. Par contre, les normes juridiques ne partagent pas ce profil
ontologique, vu qu’elles doivent avoir toujours une correspondance directe avec le monde
actuel, et que cette actualisation de ce qui est prévue par la norme affecte et modifie tant la
réalité du monde comme aussi l’état du système normatif. Deuxièmement, par l’analyse de la
notion de fiction selon Hans Vaihinger (l’auteur originalement considéré par Kelsen lors de sa
187 Pour une description du contexte général du positivisme juridique, voir BOBBIO, 1997 et BARZOTTO, 1999.
180
formulation de la notion de norme fondamentale comme fiction) on éclaire un aspect suivant
oublié par les critiques de Kelsen, à savoir, que la fiction est un élément qui n’est pas inséré
dans le Droit, qui ne fait pas partie du Droit, mais qui seulement sert au Droit, comme un outil
méthodologique, scientifique, et toujours extérieur. On se servira encore de la notion de
norme fondamentale kelsénienne comme une fiction (porteuse de contradictions) pour le
Droit, bien contraire aux normes juridiques, que ne peuvent pas avoir caractéristiques
contradictoires, pour illustrer cette approche. Dernièrement, on analysera une importante
critique concernant la négation de l’existence des normes fictives pour une défense du Droit
comme globalement fictive. On essayera de répondre a cette critique avec le même argument
utilisé antérieurement pour renforcer notre thèse en défense de la non-fictivité du Droit, en
montrant l’inadaptation des normes au caractère fictive selon la théorie artéfactuelle à cause
de la nécessité de l’actualisation dans le monde actuel de ce qui est prévue par une norme
juridique pour qu’elle puisse être appelée comme norme valide et efficace. On démontrera
aussi quelques autres points faibles de cette critique à Kelsen. Ces trois lignes principaux de
notre argumentation seront suffisantes pour le refus de la possibilité d’un traitement du Droit
et de ses normes comme fictifs.
1. Première question : quelle est la référence des normes dites « fictives » ?
Dans cette section on analysera la notion de norme fictive sous l’approche de la théorie
artéfactuelle de Amie Thomasson (1999). Cette approche est destinée surtout au domaine de
la littérature, mais on essayera de montrer justement que la norme fondamentale fictive
proposé par Kelsen (qui pourtant n’est pas une norme du Droit) a les mêmes caractéristiques
ontologiques que les objets fictifs de la littérature. C'est-à-dire, la norme fondamentale est
aussi une fiction selon la théorie artéfactuelle. On démontrera ensuite que les normes
juridiques du Droit ne s’adaptent pas à cette théorie, et que ces normes ne peuvent pas être
traitées comme fictions.
Cependant, on analysera d’abord la notion de norme fondamentale telle comme elle est
proposée par Kelsen.
1.1 La norme fondamentale : caractéristiques et fonction
181
La fonction de la norme fondamentale est de fonder et de justifier l’ordre juridique comme
un système des normes positives. Elle ne se dirige pas vers une conduite déterminée, elle ne
prescrit pas une sanction : son objet est l’ordre normatif même, comme un tout.
Toutefois, pour bien comprendre la notion de justification d’un ordre normatif, c’est
essentielle la différenciation basique entre une norme et une proposition188. Une proposition
sur la réalité du monde est dite vraie quand elle est confirmée par le moyen de notre
expérience. Mais la norme n’est pas un énoncé de l’être, elle n’est pas une proposition sur la
réalité, et ne peut pas être vérifiée dans le monde de la même manière. Tandis que la
proposition est l’issue d’un acte de pensée, la norme est l’issue d’un acte de volonté. Alors, la
norme n’est pas vraie ou fausse : elle peut être seulement valide. La notion de validité est
essentiellement liée à l’existence même de la norme : la norme est valide si elle existe dans
un ordre normatif. Dire d’une norme qu’elle n’est pas valide n’a pas de sens: si la norme n’est
pas valide, elle n’est pas une norme.
Alors, si on veut discuter le statut des normes sur un point de vue logique et métaphysique,
on doit analyser d’abord les relations entre la norme fondamentale et la Constitution.
La justification d’une norme individuelle est faite par l’existence d’une norme plus générale
dans l’ordre. C'est-à-dire, le juge est autorisé à créer une norme individuelle dans le tribunal
s’il est soutenu par l’existence d’une norme plus générale correspondante. Cette norme
générale tient sa validité justifiée par l’existence d’une norme encore plus générale, et cette
régresse continue jusqu'à la constatation de la Constitution d’un Etat189. Alors, c’est une erreur
commune de confondre la norme fondamentale avec la première Constitution d’un Etat,
parce que la validité de la propre Constitution dépend encore d’une norme «plus haute», qui
rend possible sa conformité au système comme un tout. Grâce à la présupposition de la norme
fondamentale, le contenu de la Constitution peut être considéré comme valide et obligatoire,
vu que la Constitution est déjà un document concret élaboré par les hommes, et celle-ci n’est
pas valide en elle-même. Elle est valide en appel à la norme fondamentale, qui est présupposé
comme valide et comme la condition d’existence des autres normes de l’ordre juridique. La
fonction de la norme fondamentale est, comme on a déjà dit, de donner les conditions,
188 Une discussion plus détaillée de la justification dans le domaine normatif est dans ATIENZA, 2003. 189 Pour une discussion concernant les relations entre la validité de la norme individuel et la validité de la norme général, voir CABRERA, 2000.
182
d’attribuer les pouvoirs pour la création du Droit et pour son interprétation par la science du
Droit. Kelsen (2005, p. 170) dit :
La norme fondamentale n’est pas créée dans une procédure juridique
par un organe créateur de Droit. Elle n’est pas – comme la norme
juridique positive – valide pour avoir été créée de telle et telle façon
par un acte juridique, mais elle est valide pour être présupposé
comme valide ; et elle est présupposé comme valide parce que sans
cette présupposition aucun acte humain pourrait être interprété
comme un acte juridique et, spécialement, comme un acte créateur
de Droit.190
On voit clairement par ce passage le caractère essentiellement pragmatique de la
norme fondamentale: elle doit être présupposée parce qu’il n’y a pas d’autre manière de
justifier l’ordre comme un ordre juridique légitime. Sans la norme fondamentale, la science
du Droit ne pourrait pas analyser et interpréter l’ordre juridique comme tel. C’est alors dans
le domaine scientifique de la jurisprudence que se trouve la réelle raison d’être de la norme
fondamental.
Ainsi, la probable formulation de la norme fondamentale est de cette façon liée à la
Constitution dont la validité est légitimée : les différents ordres normatifs ont des différentes
normes fondamentales, et chacune de ces normes légitime une différente Constitution. Ainsi,
dans le cas dont une Constitution est supprimée – par une révolution ou un coup d’Etat, par
exemple – la norme fondamentale est automatiquement changée, parce qu’elle se dirige vers
une nouvelle Constitution, que se substitue à l’ancienne. Avec une nouvelle Constitution, le
fondement de validité de tout cet ordre normatif doit, selon Kelsen, aussi changer.
190 Dans la version en Portugais: “A norma fundamental não é criada em um procedimento jurídico por
um órgão criador de Direito. Ela não é – como é a norma jurídica positiva – válida por ser criada de certa
maneira por um ato jurídico, mas é válida por ser pressuposta como válida; e ela é pressuposta como
válida porque sem essa pressuposição nenhum ato humano poderia ser interpretado como um ato
jurídico e, especialmente, como um ato criador de Direito.”
183
1.2 Une approche qui peut donner une réponse: la théorie artéfactuelle
Les caractéristiques particulières de la norme fondamentale nous renvoient à la théorie
artéfactuelle de la fiction. Donc, on soutient que, d’accord avec cette théorie, la norme
fondamentale conserve encore son caractère fictif, différemment des autres normes du
système juridique positif, qui ne peuvent absolument être fictives.
La théorie artéfactuelle se dirige vers le domaine littéraire, pour analyser le statut ontologique
des objets abstraits crées par les auteurs des œuvres littéraires, comme les personnages des
romans, par exemple. Si alors on applique cette théorie au contexte normatif du Droit, on
peut faire référence à la norme fondamentale comme un objet dépendant constant de la
Constitution, de la même manière que les personnages sont dépendants constants des
exemplaires publiées du roman en question. Dans le Droit, chaque Constitution doit être
soutenue par une Norme Fondamentale et, de même, aussi chaque Norme Fondamentale doit
s’adresser à une Constitution. Sauf que la norme fondamentale ne s’adresse pas seulement à
cette Constitution, mais aussi à tout l’ordre qui est en derrière. On peut faire cette relation
entre la norme fondamentale et la Constitution de manière directe parce qu’on considère
toujours un système graduel, dont la validité des normes est intra-dépendante. La
Constitution peut exister sans la norme fondamentale, mais c’est seulement avec l’existence
de cette norme que la procédure de création d’un document comme la Constitution est
justifiée juridiquement. De cette manière, on peut accéder à la Norme Fondamentale via
seulement ce document ancré dans le monde actuel, et produit par une activité humaine, qui
gagne le statut de Constitution à cause de la norme fondamentale, et que permet l’existence
du reste de l’ordre pour avoir été créé légitimement avec le soutien de cette fiction.
Dans la littérature, la relation de dépendance entre le personnage fictif et son auteur est
appelée une dépendance historique : ce l’auteur que, dans un moment donné, crée le
personnage, et cette personnage peut être dorénavant reconnu et identifié même sans la
présence de son auteur, après sa mort par exemple. Or, la relation entre la Norme
Fondamentale et son auteur est un peu plus compliquée. La norme fondamentale doit être
reconnue par la pensée juridique pour que le système soit l’objet de l’interprétation juridique.
Alors, cette interprétation est conditionnée par la présupposition de la norme. Alors, la norme
doit être supposée dans la pensée juridique toutes les fois que le système est interprété. Alors,
184
la norme fondamentale est toujours considérée par le scientifique juridique, mais ce n’est pas
lui l’auteur de la norme. Seul un membre du Droit, un sujet autorisé, peut créer une norme,
même une norme fictive dans le sens de la norme fondamentale. Alors, c’est le scientifique
juridique que reconnaît la norme comme le résultat d’un acte fictif de la volonté du sujet
compétent. C'est-à-dire, la relation entre la norme et son « auteur » est historique, mais dans
le contexte dont la norme est considérée, le contexte de la science du Droit, cette norme a
une dépendance constante en relation au scientifique qui la conçoit comme le résultat d’un
acte fictive de volonté d’autrui.
Pourtant, chaque Norme Fondamentale s’adresse à chaque différente Constitution, selon
chaque système normatif. C’est-à-dire, chaque Constitution demande sa propre fiction de
norme fondamentale, dans le moment de l’interprétation juridique. Par conséquence, la
probable formulation de la norme peut s’énoncer, selon Kelsen, comme: « On doit observer
le contenu de cette Constitution ». De cette manière, toutes les normes fondamentales
peuvent avoir la même formulation, et tout en s’adressant chacune à sa Constitution
respective (et seulement à elle). De cette manière, on peut considérer une norme
fondamentale seulement par rapport à une Constitution spécifique.
Les autres normes juridiques n’ont pas le même caractère fictif que la norme fondamentale.
Cette affirmation est beaucoup plus facile de soutenir si on considère les conceptions de
Vaihinger, par exemple, qu’on analysera ensuite. C’est d’ailleurs déjà important de noter c’est
Vaihinger l’auteur qui Kelsen fait appel toujours quand il traite des fictions. Mais, en relation
à la théorie artéfactuelle, on peut donner aussi et déjà quelques arguments pour maintenir la
thèse de que les normes juridiques positives en général ne sont pas fictives.
1.3 Une piste pour la réponse : normes juridiques doivent être aussi efficaces
En relation à cette théorie, proposé par Amie Thomasson (1999), on comprend que les normes
ont les mêmes caractéristiques que les objets fictifs (comme les personnages d’un roman
policier, par exemple) : les normes positives sont aussi comme des « abstract artifacts
dependent on literary works and created by authors » (p.148). Les normes sont objets
abstraits dépendantes du corps de loi, et elles sont créées par une personne autorisée à le
faire. Mais cela ne signifie pas que les normes sont « la même chose » encore que les objets
185
fictifs, c'est-à-dire, qu’elles sont aussi des fictions. Les normes juridiques ont quelques
caractéristiques en plus qu’empêchent cette dernière identification.
Les normes juridiques « gagnent » l’existence d’une façon toute particulière. Elles doivent être
le sens d’une acte de pensée (comme les personnages littéraires, les concepts de la science…),
mais elles doivent être aussi le sens d’un acte de volonté (comme les impératifs). En plus, cet
acte doit être objectif, c'est-à-dire, il doit être réalisé par un sujet neutre, autorisé à créer la
norme, à la rendre ainsi obligatoire à ses destinataires. Le juge crée les normes individuelles
dans le tribunal, le législateur crée les normes générales du « corps de loi »191.
Alors, les objets fictives sont issues d’actes de pensée, ou « mental states ». La norme
fondamentale est aussi le sens d’un acte de pensée « réel », et plus d’un acte de volonté fictif.
C’est à cause de cela que Kelsen se permet de dire qu’elle a des propriétés contradictoires
qu’on analysera ensuite. Or, il dit cela justement dans la phase de ses écrits dont son objectif
majeure est de bien différencier les normes des propositions192 ! Donc, cette affirmation ne
serait pas gratuite, ni naïve pour partie de Kelsen. Il sait bien qu’il ne peut pas dire d’une
norme qu’elle est contradictoire, sans compromettre tous ses conceptions pour le Droit
positif. Et pourtant il doit appeler la norme fondamentale encore de norme, parce que les
normes juridiques de l’ordre ne peuvent être justifiées que par des autres normes193.
Concernant les normes juridiques, crées par un acte de volonté objectif, on ne peut rien dire
de contradictoire, de sa vérité ou de sa fausseté. En fait, dans le contexte du positivisme
juridique, le mot « existence » a un sens tout particulier pour les normes. On ne peut ni même
les appeler « normes » qu’après la réalisation de toutes les procédures pour son existence
normative, c'est-à-dire, pour sa validité.
Alors, dans ce contexte de la théorie artéfactuelle, si on comprend la fiction comme une l’objet
d’un discours non-référentiel, on croit avoir encore des arguments pour soutenir la thèse de
que, contrairement à la Littérature, l’objet du Droit n’est pas fictif.
1.4 La réponse : la référence des normes juridiques est donnée dans son application
dans le monde actuel
191 Pour une excellente analyse de la conception de la production normative comme l’exercice d’un pouvoir chez Kelsen, voir BOBBIO, 2008. 192 Voir KELSEN, 1996. 193 Pour une critique à l’autoproduction normative de Kelsen, voir ROSS, 2003.
186
Les normes expriment une prescription, elles disent que quelque chose doit se passer dans le
monde194, et que si cette chose ne se passe pas, alors une sanction doit être appliquée. Cela
est bien différent de dire qu’une chose se passera dans le monde. Ces deux champs – celui de
l’être et celui du devoir-être – sont irréductibles, et on ne peut pas traiter les normes de la
même manière qu’on traite les énoncés sur les faits du monde195. Mais ses deux domaines
gardent encore autres types de relations, selon Kelsen. La norme doit toujours avoir un
« écho » dans le monde vers où elle se dirige. Cela veut dire que la norme doit être efficace,
elle doit être obéit par ces destinataires, ou encore la sanction qu’elle prescrit doit être
appliqué par les organes prévues. Cela ne veut pas dire que la norme peut être réduite ou
encore déduite à partir de cette conduite factuelle. Quand une norme est appliquée ou
observée, les deux domaines expriment le même contenu, concernant une conduite
déterminé, et alors sont les contenus que sont correspondants, et pas les modes
d’expression196.
C’est dans un sens « pratique » que la norme a des rapports avec le monde. Cette affirmation
ne permet encore de dire qu’on pourrait traiter les normes comme des énoncés : Kelsen veut
dire simplement qu’il est inutile et même irrationnel créer une norme qui ne peut pas être
observé par personne. Les référents de ces normes dans le monde actuel sont ainsi des
conduites qui ont été ou qui seront conformes à ces mêmes normes, c'est-à-dire, le contenu
de la norme, qui trouve une correspondance dans le monde. De cette manière, l’application
d’une loi révèle en soi sa vérification dans le monde actuel. De la même manière, les normes
qui autorisent un sujet à effectuer une opération interne du Droit sont constatées lors de la
réalisation de cette opération. Clairement, cela n’est pas la même chose que dire qu’un objet
fictif, par exemple, le personnage Leopold Bloom, est vérifié dans le monde au moment des
célébrations du Bloomsday à Dublin.
Bref, on peut dire que toutes les normes sont en fait limitées par la modalité du « possible »,
mais l’important c’est que le « possible » du Droit positif doit toujours être actualisé dans un
certain moment. Un bon exemple c’est le cas de la « loi sèche », connue aux Etas Unis des
années 20. Au Brésil, et aussi dans autres pays, une loi similaire est destinée aux conducteurs
194 La norme est plus qu’une commande, selon Kelsen : elle a aussi un caractère obligatoire à cause d’être produite par le juge. Pour une discussion plus complète, voir KELSEN, 2005. 195 Sur le sophisme naturaliste, voir HUME, 2000. 196 Voir la discussion sur le substrat indifférent au mode, voir KELSEN, 1996. Pour une autre approche sur cette même question, voir HARE, 1996.
187
de voitures. Pourtant, la manque d’inspection de la part des policiers a répercuté dans
l’observation de la norme pour les citoyens, et on peut dire qu’aujourd’hui la norme a presque
perdu sa validité à cause de la perde de son efficace. Cependant, la « loi sèche » original des
Etas Unis, qui a prohibé la fabrication, le commerce, le transport, l’importation et l’exportation
de boissons alcoolisées a perdu complètement sa validité par abrogation, par un acte
juridique, à 1933, à cause aussi de la manque d’efficace. On peut dire, donc, que le domaine
normatif exerce une interférence sur le monde actuel : le monde actuel est changé à partir
des normes, mas le domaine normatif est aussi changé à partir des faits. Toutefois, on ne peut
dire le même des objets fictifs, vue que la fiction est essentiellement détachée de la réalité du
monde actuel. Même si une histoire peut parler avec beaucoup de vraisemblance sur un
événement réel, cette histoire ne se destine pas à changer quelque chose dans la réalité. Cette
dynamique concernant la relation constante entre les faits et les normes est une
caractéristique qu’empêche l’insertion de l’élément de la fiction dans le Droit. Le fait essentiel
c’est que, sans l’actualisation de ce qui est possible selon le devoir-être dans l’être, dans le
monde actuel, la norme perd son efficace, et alors elle perd aussi sa validité, c'est-à-dire, elle
n’existe plus. Alors, la notion de validité normative comme ayant besoin d’un minimum
d’efficacité dans le monde actuel rejette pour soi-même le caractère fictif. Pourtant, si Kelsen
conçoit la norme fondamentale encore comme une fiction, c’est parce qu’elle n’est pas une
norme dans ce sens-là.
1.5 La norme fondamentale ne peut pas être efficace : elle est une norme « impossible »
Autre point important qui doit être brièvement mentionnée c’est le fait que, selon Kelsen, le
Droit doit être pleinement séparé de la science que le décrit. Cette affirmation répercute
directement sur la question des possibilités d’application de la logique traditionnelle aux
normes. Comme la norme est toujours le sens d’un acte de volonté (objectif), tandis que la
proposition est le sens d’un acte de pensée, Kelsen trouve déjà un argument solide pour
attester l’abîme méthodologique entre les domaines de l’être et du devoir-être, irréductibles
et incomparables. Le domaine du Droit est le domaine prescriptif des normes valides. Le
domaine de la Science du Droit (le domaine de la théorie proposée par Kelsen) c’est le
domaine des descriptions sur le Droit, des descriptions vraies ou fausses sur les normes
188
valides. Alors, l’axiome qui pose la différenciation entre ces deux domaines est la base pour la
négation kelsénienne de l’application au Droit (aux normes) de la Logique Traditionnelle.
Dans ce contexte, Kelsen reconnait déjà quelques problèmes (simplement apparents, selon
lui-même) qui se relient avec la fonction et l’importance de la norme fondamentale. Par
exemple, si on compare les deux énoncés suivants, du domaine moral : « Tu assisteras ton
semblable quand il en aura besoin » et « Tu mentiras toutes les fois que tu le jugeras utile »,
la théorie de Kelsen indique que seulement le premier énoncé exprime une norme valide.
Cette thèse se justifie parce que la recherche pour le fondement de la validité normative,
selon Kelsen, vise toujours l’existence d’une norme antérieure, plus générale, et cette
recherche doit s’arrêter dans une norme qui n’a pas elle-même le besoin d’être justifiée, qui
tient sa validité comme évidente. La validité de cette norme ne se fonde pas dans la validité
d’une norme encore plus générale, ce qui signifierait un regressus ad infinitum. Dans notre
exemple, la norme « Tu assisteras ton semblable quand il en aura besoin » fait partie d’un
encadrement normatif qui peut nous guider jusqu’à une norme complexe et global comme :
« Aime ton prochain comme toi-même ». Autrement, l’impératif « Tu mentiras toutes les fois
qui tu le jugeras utile » ne peut pas être reconnue comme faisant partie d’un encadrement
normatif, qui doit remonter à une norme de validité évidente et pleinement reconnue.
Autant que le point culminant de la recherche pour la justification juridique, la norme
fondamentale ne peut pas avoir le même caractère que toutes les autres normes de l’ordre.
Si la norme serait formulée de manière standard, elle aurait aussi le besoin d’être elle-même
justifiée. Donc, la norme fondamentale gagne un statut différencié, elle est placée dans un
niveau ‘panoramique’ en relation au tout de l’ordre : pour justifier l’ensemble de normes en
donnant l’unité du système, la norme fondamentale est seulement supposée par le sujet
compétent. Elle est le sens d’un acte de volonté fictif. Cela veut dire que cette norme est
conçue comme consciemment crée par le sujet compétent comme une norme fictive, avec un
propos particulier bien défini. Cette norme est impossible parce que sa propre production
indique une erreur au moment de la production d’une norme : la norme fondamentale ne
reçoit pas la légitimation d’une norme plus « haute », le juge n’est pas autorisé à la produire.
Dans ce sens elle doit être considérée une fiction, parce qu’il est impossible obéir ou appliquer
une norme que ne fait pas partie d’un système, qui est seulement supposé par son auteur
avec un propos méthodologique extérieur aux démarches du Droit. La norme fondamentale
n’est pas une norme à obéir, mais seulement une apparatus qu’on regarde comme norme
(fictive) pour pouvoir considérer l’ordre normatif comme un ensemble cohérent et possible
189
d’être interprété. Ainsi, le rôle principal joué par la norme fondamentale est de légitimer la
possibilité de la création normative. C’est-à-dire, le fait de pouvoir arriver à une norme
fondamentale dans la recherche pour la justification normative atteste que toutes les
conditions pour la création normative ont été remplies.
On verra ensuite comment Kelsen a caractérisé cette norme, dans deux phases dans ses écrits.
Malgré la reformulation de la notion de norme fondamentale par l’auteur, un aspect demeure
inchangé : la norme fondamentale ne fait pas partie du système juridique avec des autres
normes. Elle possède des caractéristiques beaucoup particulières pour être mise avec ces
autres normes. Ces caractéristiques la relient surtout avec la science du Droit, et pas avec son
objet, les normes juridiques.
2. Deuxième question : comment une norme peut-elle être contradictoire ?
On arrive maintenant à la considération d’une autre approche sur la notion de norme
fondamentale, à savoir, la philosophie du « comme si » de Hans Vaihinger. Cette approche est
la base pour la conception de norme fondamentale comme une fiction de la Science du Droit
selon Kelsen, et elle est signalée par la présence des contradictions dans la formulation de la
notion de norme fondamentale. Pourtant, on analysera d’abord la première conception de
norme fondamentale comme hypothèse, sous l’influence de Kant.
2.1 Norme fondamentale comme hypothèse : l’influence de Kant
La norme fondamentale n’est pas une norme positivée, elle n’est pas établie par un individu,
car personne n’est autorisé à la produire de cette façon (il n’y a pas d’autre norme plus haute
qui autorise quiconque à la produire). Elle est une norme présupposée dans la pensée
juridique. Mais la norme fondamentale n’est pas seulement le sens d’un acte de pensée : elle
est le sens aussi d’un acte de volonté: un acte de volonté fictif.
Dans la Théorie Pure du Droit (1962, 2a éd.), Kelsen révèle une certaine influence de Kant,
quand il fait une comparaison entre la justification de la norme fondamentale et la justification
190
des concepts purs de la compréhension197. Dans le longue passage ci-dessous, on reconnaîtra
la première position de Kelsen sur les caractéristiques de la norme fondamentale, qui la relient
à la philosophie transcendantale kantienne, en assumant la norme fondamentale comme
hypothèse logique-transcendantale pour l’interprétation des normes de l’ordre par la science
du Droit :
En tant que l’hypothèse d’une certaine norme fondamentale permet seule de reconnaître à
l’acte ou au fait constituants, et aux actes ou faits réalisés conformément à cette Constitution,
la signification objective qui concorde avec leur signification subjective, c’est-à-dire de leur
reconnaître le caractère de normes juridiques objectivement valables, on peut dire que la
norme fondamentale que présente la science du droit198 est - s’il est permis d’utiliser par
analogie un concept de la théorie kantienne de la connaissance – la condition logique
transcendantale de cette interprétation. De même que Kant pose la question de savoir
comment une interprétation des faits donnés à nos sens exempte de toute métaphysique est
possible dans les lois naturelles formulées par les sciences de la nature, de même la théorie
pure du droit pose la question de savoir comment est possible une interprétation de la
signification de certains faits comme une système de normes juridiques objectivement
valables, descriptibles dans des propositions de droit, une interprétation qui ne recoure pas à
des autorités métajuridiques telles que Dieu ou la Nature.
A cette question de théorie de la connaissance, la théorie pure du Droit répond : une telle
interprétation est possible à la condition de supposer la norme fondamentale, « on doit se
conduire comme la Constitution le prévoit », c’est-à-dire de la façon qui correspond au sens
subjectif de l’acte de volonté constituant, aux prescriptions ou commandements du
législateur constituant.
La norme fondamentale du Droit, en tant que supposition, peut être comprise comme une
hypothèse. Néanmoins, un détail interdit cette façon de la comprendre: cette norme garde
dans sa propre formulation des éléments qui contredisent la propre théorie dont elle est issue.
La norme fondamentale n’est pas une simple supposition, elle est aussi un type de formulation
normative en contradiction avec les conditions de sa propre production. Sa « production » est
197 Pour des analyses de l’influence par Kant dans la théorie de Kelsen, voir KAUFMANN, 1992 et GOYARD-FABRE, 1993. 198 On mettre en gras cette partie du passage pour mettre en évidence le fait que Kelsen lui-même situe la norme fondamentale dans le contexte de la science, et pas dans le Droit.
191
une infraction à la production normative. Autrement dit, la norme fondamentale n’est pas
une hypothèse parce qu’elle n’est pas ni même possible dans l’ordre juridique.
2.2 La norme fondamentale comme fiction : l’influence de Vaihinger
Dans l’ouvrage Théorie Générale des Normes (1979), Kelsen ira jusqu’à abandonner et même
nier l’identification de la norme fondamentale avec une hypothèse. Dans cette nouvelle
phase, il revisite la notion de norme fondamentale sous l’éclairage de la philosophie du
« comme si » de Hans Vaihinger :
Comme il ressort de ce qui précède, la norme fondamentale d’un ordre moral
ou juridique positif n’est pas une norme positive, mais une norme simplement
pensée, c’est-à-dire une norme fictive, la signification d’un acte de volonté
non pas réel, mais fictif. En tant que telle, elle est une fiction véritable au sens
de la « philosophie du comme si » de Vaihinger, une fiction caractérisé par le
fait que non seulement elle contredit la réalité, mais aussi qu’elle est
contradictoire en soi. En effet, l’hypothèse d’une norme fondamentale – par
exemple la norme fondamentale d’un ordre moral religieux : « On doit obéir
aux commandements de Dieu » ou la norme fondamentale d’un ordre
juridique : « On doit se comporter comme le détermine la Constitution
historiquement la première » - contredit non seulement la réalité, car aucune
norme de ce genre n’existe en tant que signification d’un acte de volonté réel,
mais aussi elle est contradictoire en soi, car elle représente l’habilitation
d’une autorité juridique ou morale suprême, et part ainsi de l’idée d’une
autorité (toutefois purement fictive) plus haute encore que cette autorité.
Une fiction est, selon Vaihinger, un expédient de la pensée dont on se sert si
l’on ne peut pas atteindre le but de la pensée avec les donnés dont on dispose.
Le but de la pensée de la norme fondamentale est le fondement de la validité
des normes constituant un ordre moral ou juridique positif, c’est-à-dire
l’interprétation de la signification subjective des actes posant ces normes
comme leur signification objective (c’est-à-dire comme normes valides) et les
actes concernés comme des actes posant de normes. Ce but ne peut être
atteint que par la voie d’une fiction. C’est pourquoi il convient de remarquer
192
que la norme fondamentale, au sens de la « philosophie du comme si » de
Vaihinger n’est pas une hypothèse – comme je l’ai moi-même quelquefois
caractérisé –, mais une fiction qui se distingue de l’hypothèse par le fait
qu’elle est accompagnée ou doit être accompagnée de la conscience que la
réalité ne lui est pas conforme. (KELSEN, 1996, p.344-345)
Quand les critiques parlent de la norme fondamentale kelsénienne comme fiction, rarement
ils ont dans la tête la notion de fiction de la manière que Kelsen l’a caractérisée, c'est à dire,
une fiction dans le sens de la philosophie de Vaihinger. De cette manière, c’est commun
d’entendre parler de la norme fondamentale comme une fiction dans le sens de simplement
ne pas être un objet « réel ». Cependant, on est bien d’accord que la norme fondamentale est
un objet abstrait, comme les sont aussi toutes les autres normes. Mais cela n’implique pas
que ces normes soient aussi des fictions. Au moins, cela n’est pas de tout d’accord avec ce que
Kelsen a proposé. Sa conception de fiction s’engage essentiellement avec l’élément de la
contradiction et, pourtant, il place la norme fondamentale au coté des énoncés d’être, et pas
au milieu des normes du Droit.
2.3 La piste pour la réponse : on ne peut pas observer une norme contradictoire
La norme fondamentale comme fiction selon l’interprétation de Kelsen nous contraint à
reconnaître l’élément de la contradiction, en ce sens qu’elle n’existe pas dans la réalité de
l’ordre juridique. C’est à dire, la norme est contradictoire à la réalité parce qu’elle n’est pas
même une norme formellement présente dans l’ordre juridique, puisqu’elle n’avait pas été
créée au moyen d’un acte de volonté réel, un acte subjectif dans le monde des faits et qui a
été objectivé par la figure du législateur. De plus, Kelsen dit que la norme est « contradictoire
en soi », parce que la formulation même de la norme fait appel à une autorité plus haute que
celle du législateur. Cette autorité servirait à formuler la norme comme telle. Ainsi, la présence
des contradictions empêche-t-elle l’utilisation du terme « hypothèse » pour caractériser la
norme fondamentale. En effet, une hypothèse, dans une théorie scientifique, ne peut pas
contredire la réalité. En revanche, elle doit partir des présuppositions constatables, pour
rechercher un résultat déterminé, qui viendra à confirmer ou réfuter la présupposition. Même
si une hypothèse peut avoir une caractéristique apparemment fictionnelle (dans le sens
193
qu’elle n’est pas encore présente dans la réalité, mais seulement supposée présente pour
qu’on puisse vérifier comment le monde serait si l’hypothèse était soutenue), elle ne s’engage
pas avec des contradictions. Autrement dit, Kelsen doit éliminer la notion d’hypothèse parce
que l’élaboration de la norme fondamentale se donne avec la conscience même, pour partie
de son auteur, de ce que cette « opération » n’est pas en accord avec des principes de la
production normative. Ainsi, on constate une espèce d’exception pendant la formulation de
la norme comme fiction, un « attentat » aux principes positivistes pour la production
normative. Mais son but est justifié par les nécessités théoriques de clôture et de légitimation
de l’ordre juridique, attestées par la science du Droit.
J. Loewenberg (1912, p.717) démontre de manière claire l’opposition entre la notion de fiction
soutenue par Vaihinger, face à la notion d’hypothèse, et face aussi à des autres notions de
fiction, comme celles de la théorie artéfactuelle, par exemple :
Fictions, in Vaihinger’s usage, are not identical with figments, such as centaur
or fairy, nor are they hypotheses capable of verification. They are deliberate
devices (Kunstgriffe) on the part of thought for the practical purpose of
successful orientation in and perfect control over the environment.
Theoretically they are absolutely valueless. Applied with a knowledge of their
fictitious character, they will lead to the intended practical results, but used
as hypotheses, they must necessarily create confusion and false theories, for
a fiction is defined as that which is both contradictory in itself and which has
no correspondence with reality (sensational flux).
Ce passage illustre l’aspect d’utilité pratique de la fiction selon Vaihinger, et cette
caractéristique est de même central pour l’identification de la fiction selon cet auteur. Ainsi,
une Logique pour étudier la fiction de norme fondamentale est faisable et même convenable
dans la théorie kelsénienne. L’étude logique de la notion de cette norme comme fiction ne
représente pas une présence de la Logique dans le Droit. Cela serait vivement condamné par
Kelsen. Ce traitement logique se situe dans la théorie du Droit, dans le domaine descriptif de
la science du Droit, dont on peut simplement constater l’existence de la norme fondamentale
comme un objet fictif. Donc, la norme fondamentale est une fiction pour le Droit, une fiction
placée dans la Théorie du Droit, qui est une science composé par propositions, et pas pour
normes.
194
Malgré l’insistance de Kelsen en dire que la norme fondamentale n’a pas le même statut que
les normes positives du Droit, beaucoup de critiques considèrent les derniers écrits de Kelsen
comme un exemple de "fictionnelisme" du Droit. On présentera ensuite les arguments contre
cette interprétation.
2.4 Réponse : La norme fondamentale est contradictoire parce qu’elle ne prescrit
pas un devoir-être réel
Selon le vocabulaire juridique, la fiction « constitue une méthode de raisonnement, un
procédé de technique juridique ‘consistant en supposer un fait ou une situation différents de
la réalité en vue de produire un effet de Droit’ » (WICKER, 1997, p.9). Cette définition est en
accord avec les limites du propos instrumental de la norme fondamentale dans la théorie
kelsénienne : le scientifique juridique, en analysant un ordre pour vérifier si l’ensemble des
normes constitue un corps normatif valide, se sert de l’outil méthodologique de la fiction de
norme fondamentale pour confirmer la clôture de l’ensemble, et justifier de cette manière la
validité de l’ordre comme un tout. Le scientifique avance dans le processus de justification
jusqu’à la Constitution. Ensuite, il assume l’existence fictive d’une norme fondamentale, qui
donne la signification objective des actes juridiques, c’est à dire, la validité même des normes
dans l’ordre considérée. C’est crucial de percevoir que la norme fondamentale est considérée
dans le moment de l’analyse, de l’interprétation. La norme fondamentale est présupposée
dans la pensée juridique, dans le moment dont on fournit le caractère objectif aux actes de
volontés subjectifs, quand on interprète le sens de l’acte de volonté subjectif comme aussi
objectif. Cette procédure est dans un niveau plus haut que ce de la production du Droit. On
est dans le niveau de la justification du Droit, de l’interprétation juridique. Alors, quand Kelsen
approche la notion de fiction, il reste toujours en conformité avec ce que Vaihinger prétende :
les deux auteurs mettent en relief la notion de fiction comme un apparatus qui sert à une
théorie scientifique, que nous aide dans la compréhension de quelque chose dans le monde,
par le moyen des suppositions qui contredisent ces faits, mais que se justifient en raison de
sa utilité pratique.
Pour appartenir à ce domaine théorique, la norme n’est pas considérée par les individus
soumis à l’ordre normatif. Même les avocats ne considèrent pas la question de la norme
fondamentale comme essentielle ou même comme porteuse d’une influence dans la pratique
195
du Droit. Pour la pratique quotidienne du Droit, la fin de l’ordre normatif se situe dans la
Constitution. C’est le scientifique du Droit qui doit reconnaître l’importance théorique de la
norme fondamentale, comme condition essentielle de reconnaissance de l’ordre normatif
entant qu’objet soumis à l’investigation scientifique. La norme fondamentale intéresse le
scientifique du Droit parce que c’est elle qui donne l’unité, l’encadrement objectivement
ordonné des différentes normes reconnues comme valides. C’est elle qui permet de
reconnaître que les normes sont comme un tout valides et appartenant à un seul ordre
normative.
3. Troisième question : une norme appliquée est-elle simplement une norme
possible ?
Dans cette troisième section, on analysera une important critique à la notion de norme
fondamentale comme fiction. On essayera de préserver la notion de norme fondamentale
comme une fiction pour le Droit, et de faire ressortir les éléments que démontrent les
différences entre cette norme et les normes juridiques quant au caractère fictif. Le principal
point de la critique c’est le fait que le domaine normatif serait limité par la modalité du
« possible », et alors il serait fictif comme un tout. Notre réponse se dirige vers la constatation
de que les normes ont des relations strictes avec la réalité vers laquelle elles se dirigent, et
que la norme fondamentale ne se comporte pas de la même manière. On essayera de montrer
les interférences directes entre l’application d’une norme et les répercussions dans le monde
actuel pour soutenir que l’application d’une norme peut être vue comme une instanciation de
cette norme. La norme fondamentale n’est jamais appliquée dans ce sens-là : elle est
seulement constatée comme une fiction pour la justification normative.
3. 1 Critiques à Kelsen
Il y a actuellement des discussions très enflammées concernant le statut de la norme
fondamentale chez Kelsen. Une critique importante sur l’approche de Kelsen concernant la
norme fondamentale comme fiction est faite par le professeur Otto Pfersmann (2004, p.52-
53) :
196
Kelsen a raison de dire que la Norme fondamentale, c’est-à-dire celle qui
fonde la validité normative des autres normes du système en question, est
une fiction puisqu’elle n’existe pas en tant que fait naturel, sauf que cela ne
vaut pas seulement pour la Norme fondamentale mais pour toutes les normes
d’un système. Kelsen a encore raison en tant qu’il considère la Norme
fondamentale comme un instrument cognitif utile en tant qu’il permet de
constituer un champ de connaissances, par exemple le droit de l’Union
européenne ou l’ordre juridique français. Mais Kelsen a tort, lorsqu’il fait de
la Norme fondamentale quelque chose de doublement « faux » au sens de
Vaihinger, c’est-à-dire qui contredit la réalité (puisqu’aucun acte ne l’a
produite) et qui est contradictoire en elle-même puisque cette construction
suppose qu’il existerait une autorité encore plus élevée que l’autorité
juridique suprême qui aurait éditée cette habilitation. Certes, il serait faux de
dire qu’il existe une autorité qui aurait édicté une Norme Fondamentale et il
serait doublement faux de penser qu’il existe une autorité morale sur-
suprême qui aurait autorisé une autorité juridique suprême à agir de la sorte.
Mais justement, ce n’est pas ce que dit le théoricien qui admet pour les
besoins de la cause que les énoncés que l’on trouve dans les documents
constitutionnels d’un ordre juridique soient considérés comme des
formulations de normes valides. Il ne suppose nullement qu’il aurait existé un
acte ayant habilité les constituants à adopter la constitution ni qu’il y aurait
une ou plusieurs autorités entre lesquelles il y aurait par ailleurs un rapport
hiérarchique d’habilitation. Ce qui est strictement faux au sens de Vaihinger
serait l’admission qu’il existerait des normes qui auraient en tant qui telles
une force obligatoire. Mais comme Vaihinger et Kelsen le disent eux-mêmes,
il s’agit ici de « constructions auxiliaires de la pensée » (Denkbehelfe) : la
fiction porte sur la validité et non sur une conception aberrante des
fondements naturels de la validité.
Cette critique de Pfersmann à la norme fondamentale comme fiction est pourtant dépendent
de son autre conception pour le Droit, à savoir, que le Droit comme un tout serait fictionnel,
pour la raison que les normes, autant que « valides », ne sont pas liées au monde des faits, de
197
la réalité. Les normes sont fictionnelles, selon Pfersmann, parce qu’elles sont signalées par la
modalité du possible.
Kelsen, par sa fois, soutient que la norme fondamentale est une fiction pour être justement
impossible dans l’ordre juridique. Les normes standards, celles que règlent la conduite
humaine, elles ne peuvent jamais être impossibles selon Kelsen. Selon l’auteur, il n’y a pas de
sens dans la création d’une norme qui ne peut pas être observé par les destinataires. La
norme, pour être valide, doit être aussi, dans un minimum, efficace. C'est-à-dire, la conduite
prescrite, le délit réglé ou la sanction fixée doivent tous pouvoir être réalisés dans le monde.
En plus de la possibilité de réalisation, cette réalisation doit vraiment avoir lieu dans un
moment donné.
Le fait de concevoir le Droit comme fictif comme un tout c’est l’unique condition qui permet
à Pfersmann d’affirmer que la norme fondamentale est « quelque chose de doublement
‘faux’ ». Pfersmann, au contraire de Kelsen, admet seulement la notion de fiction à propos de
la validité de la norme fondamentale (et de toutes les autres normes), et pas la fiction qui
s’engage avec des relations de pouvoir, c’est-à-dire, d’autorisation pour la création des
normes.
3.2 Piste pour la réponse : sans application dans le monde, la norme perd son existence
Pour Kelsen, la validité d’une norme est dans le domaine du devoir-être, et ne peut pas se
réduire au domaine de l’être. Mais Kelsen n’admet pas que la validité soit complètement
détachée de la réalité : un ordre juridique valide doit être au minimum efficace, c’est-à-dire,
au minimum respecté ou observé. Une norme ne peut jamais être « fausse » comme le dit
Pfersmann : si la norme est valide, elle existe, elle est présente dans l’ordre. C’est à cause de
cette affirmation de Kelsen qu’on se permet de dire que la norme fondamentale n’est pas au
même niveau que les autres normes. Elle se distingue parce qu’elle existe seulement à cause
d’un propos spécifique : la justification de l’ordre (au moyen de la clôture) et donc la
connaissance de l’ordre, par le scientifique, comme un corps normatif significatif. Elle n’est
pas de même nature que les autres normes, et avec une « aggravation » du caractère fictif. La
norme fondamentale n’a rien à voir avec la réalité (ou même, elle a une relation contradictoire
avec la réalité), au contraire de certaines normes qui ont pour objectif de régler la conduite
humaine. Pour ces normes, un minimum d’efficace est exigé.
198
C’est important mentionner que la thèse soutenue dans ce texte a des ambitions plus fortes,
qui malheureusement on ne peut pas développer suffisamment ici. On soutient une approche
encore plus intégral sur la notion de fiction et le Droit, à savoir, que les normes n’ont pas le
caractère fictif de la même manière que un personnage littéraire est fictif. Le caractère fictif
d’une norme est seulement éventuel ou contingent, dans le sens que une même norme dite
fictive, peut être simplement reformulée comme non-fictive sans aucune altération dans son
sens. Or, on peut attester l’existence des normes dites fictives depuis l’Antiquité, et surtout
avec des exemples du Droit Romain199. La norme dite fictive que dit, par exemple, qu’un fils
adoptif doit être considéré comme s’il était un fils légitime, telle norme pourrait être expresse,
par exemple, en disant que le fils adoptif doit avoir les mêmes droits et devoirs que le fils
légitime. La fiction est, de cette manière, une simple une façon d’expression entre autres ou,
comme on a déjà dit, une technique juridique. On ne peut pas dire le même concernant un
objet fictif comme le Minotaure ou Sherlock Holmes ou le carré rond : Sherlock Holmes ne
peut exister que comme le personnage fictif crée par Conan Doyle. Mais la norme dite fictive
peut encore exister comme norme, en s’adressant aux mêmes destinataires et avec le même
sens, sous une autre formulation normative. La compréhension de la fiction comme technique
juridique est déjà en soi-même une négation de l’existence des normes fictives. Le propre
Pfersmann accepte cette conception quand il nie l’existence des normes fictives pour
défendre la thèse de la fictionnalité du Droit comme un tout. On est bien d’accord, par
exemple, avec l’analyse de Pfersmann de la « norme fictive » selon laquelle « Seules les
femmes ont accès à cette piscine, le maître-nageur est une femme au sens de ce règlement ».
Pfersmann (2004, p.54) dit :
Dans l’exemple de la piscine, il n’est pas dit qu’un homme est une femme,
mais simplement que, au cas où il existe un maître-nageur du sexe masculin,
cette personne doit être traitée comme une personne du sexe féminin au sens
du règlement de la piscine de N***, en d’autres termes, il lui est permis de
pénétrer dans la piscine. Le même résultat aurait été obtenu, si le règlement
avait dit :
N : Seules les femmes sont admises dans cette piscine, à l’exception du
maître-nageur en charge de la sécurité des personnes et des installations.
199 Voir GUERRIER, 2004.
199
Or, il s’agit simplement d’une autre formulation de la même norme utilisant
une autre technique.
On est complètement d’accord avec cette approche sur les normes dites fictives. Voir la fiction
comme une technique signifie la voir comme un outil de la Science du Droit pour le Droit, de
la même manière qu’elle est un outil pour la Physique ou la Chimique.
Mais Pfersmann dévie de cette approche quand il considère le Droit comme entièrement fictif.
Comme on a déjà montré, même si le Droit est limité par le possible, il doit avoir une
correspondance avec le monde, dont le possible est actualisé dans un moment donné. C’est
ainsi nécessaire qu’il y ait dans le monde au moins une actualisation de chaque norme, par le
moyen de l’observation de la conduite prescrite ou l’application de la sanction fixée, ou encore
de l’accomplissement d’une fonction juridique prescrite. Si on ne obtient pas cela, alors la
norme n’existe plus, ce que montre l’existence d’une interférence de l’actuel sur le possible,
de les faits sur les normes, et non pas seulement des normes sur les faits.
3.3 Réponse : l’application d’une norme révèle une actualisation nécessaire du
« possible » du Droit
Dans le contexte du positivisme de Kelsen, quand l’auteur dit que la norme fondamentale est
une fiction, il parle donc de la fiction dans le sens dont Vaihinger la conçoit. La norme
fondamentale, et elle seulement, possède des contradictions dans sa définition même, parce
qu’elle n’est pas le sens d’un acte de volonté réel. Ce caractère nous permet de traiter la
norme avec des prédicats destinés à des propositions. Une norme positive ne peut pas avoir
de contradictions, parce qu’elle est seulement valide, et pas vraie ou fausse. Autrement dit,
la norme fondamentale est surtout une construction théorique, une « construction auxiliaire
de la pensée », et se destine aux propos théoriques de la science du Droit. Dans ce sens on
doit comprendre la norme fondamentale comme une fiction pour le Droit, parce que le Droit
même ne peut pas avoir des normes fictives, marquées par des contradictions, selon les
conceptions de Vaihinger et Kelsen. La norme fondamentale ne se dirige vers la réalité des
conduites humaines, elle se dirige vers l’intégrité de l’ordre normatif. Une norme juridique
doit être observée ou appliquée dans la réalité du monde pour continuer à être une norme,
200
mais la norme fondamentale se destine vers la validité de l’ordre comme un tout, et n’est pas
« appliquée » de la même façon. Une norme fondamentale n’a pas cette même relation avec
le monde actuel, sa validité ne dépend pas de son efficacité. Sa validité est une fiction, et se
trouve dans le champ de l’impossible.
Conclusion
L’immense travail de Kelsen pour construire une théorie « pure » pour le Droit (et même pour
le domaine normatif en général) a rencontré parfois des problèmes internes le long de ses
écrits. Kelsen lui-même n’a jamais hésité à les révéler et à corriger ses thèses. Il y a même eu
des moments où Kelsen a changé radicalement ses approches sur des problèmes pertinents
de la théorie du Droit. C’est à cause de cela que, en relation avec différents notions ou
positions kelséniennes, on peut bien identifier des « phases » dans la pensée de l’auteur.
De même pour la notion de norme fondamentale. Cette notion, probablement la plus discutée
et attaquée par les critiques, est analysée et défendue ici comme une fiction pour le Droit. En
plus, on considère aussi que dans le Droit positif il n’y a pas de place pour les fictions.
Dans la présente analyse des caractéristiques particulières de la norme fondamentale, un des
aspects les plus importants c’été la différenciation entre las normes et les propositions que
les décrivent. Cette différence nous a permis de reconnaître dans la norme fondamentale (en
accord avec la lecture kelsénienne de Vaihinger) encore quelques caractéristiques des
propositions, comme la contradiction, vue que la norme fondamentale est le résultat d’un
acte de pensée et, en plus, d’un acte de volonté fictif. Avec cette perspective, on a considéré
alors la notion de fiction sous l’approche de la théorie artéfactuelle, pour voir que la notion
de norme fondamentale conserve son statut fictive, et que cette statut n’est pas partagé par
les autres normes, qui doivent être aussi efficaces pour être valides. Après, on a analysé la
notion de norme fondamentale comme hypothèse. En fait, si on considère l’analyse de cette
norme en tant que simple présupposition du législateur, comme élément qui n’est pas présent
dans l’ordre juridique, on peut en fait noter une similitude avec la notion d’hypothèse.
Toutefois, cette identification est tout à fait impropre, à cause de l’existence des éléments
contradictoires qui composent la notion de norme fondamentale : elle possède des
caractéristiques contradictoires avec sa propre formulation. Le constat de cette erreur a
conduit Kelsen à la théorie vaihingerienne du « comme si », et à la conception finale de norme
201
fondamentale comme fiction. On a constaté que l’approche de Kelsen de la norme
fondamentale comme porteuse de contradictions, et insérée dans une science avec un propos
pratique bien défini, est « autorisé » par la théorie de Vaihinger. C'est-à-dire, on peut
considérer seulement la norme fondamentale comme une fiction légitime selon la théorie de
Vaihinger.
Cependant, cette approche est actuellement contestée par plusieurs critiques. D’entre eux,
on a analysé particulièrement la critique d’Otto Pfersmann. Cette critique est conditionnée
par la thèse de son propre auteur, selon laquelle le Droit est comme un tout fictif. Pfersmann
conçoit la norme fondamentale comme fictionnel tout autant que toutes les autres normes
juridiques, et alors la propre notion de norme fondamentale comme le point culminant de la
recherche pour la justification juridique disparaît. On a essayé de montrer que le « possible »
du Droit doit avoir une actualisation dans le monde actuel, par le moyen de l’application
normative, pour que la norme conserve sa validité. Sans cette application dans le monde, la
norme n’existe plus. Alors, la norme fondamentale n’est pas destinée à régler les conduites
dans le monde actuel, et ne peut pas être appliqué de cette façon, vue qu’elle est une norme
impossible, c'est-à-dire, que sa propre validité est fictive.
Finalement, on conclut que la notion de norme fondamentale comme fiction est bien
appropriée aux propos de la théorie kelsénienne pour le Droit, et qu’elle seulement réussit
dans son objective de justifier le système juridique pour être « séparé » du Droit, pour lui avoir
comme son objet. On soutient en plus qu’aucune norme de l’ordre positif peut-elle être une
fiction, sous aucun approche: si on comprend la fiction comme manque de référence
(Thomasson), on s’appuie sur la nécessité de l’efficace normative; si on comprend cette notion
comme un outil méthodologique qui présente des caractéristiques contradictoires pour
expliquer par supposition ce que ne peut pas être expliqué dans la réalité (Vaihinger), on
s’appuie sur le fait que les normes juridiques ne peuvent pas avoir des contradictions. La
constatation de la présence des contradictions dans la formulation de la norme fondamentale,
parfois considéré problématique, est en fait précise, et cette présence ne viole pas les
processus scientifiques qui sont réalisés par la norme. Justement, les contradictions présentes
dans la formulation de la norme en tant que fictive permettent de légitimer des démarches
juridiques au niveau de la Science du Droit, pour la justification de la validité de l’ordre
normatif comme un tout. Autrement dit, c’est essentiellement la formulation de la norme
fondamentale comme une fiction qui rend possible l’existence même d’un système juridique
positif légitime.
202
Références Bibliographiques:
ATIENZA, M. [ 2003] As razões do Direito. Teorias da Argumentação Jurídica. Tradução
de Maria Cristina Guimarães Cupertino. São Paulo: Landy Livraria e Distribuidora Ltda.
BARZOTTO, L. F. [1999] O Positivismo Jurídico Contemporâneo. Uma introdução a
Kelsen, Ross e Hart. São Leopoldo: UNISINOS.
BOBBIO, N. [2008] Direito e Poder. Tradução Nilson Moulin. São Paulo: Editora UNESP.
BOBBIO, N. [1997] Teoría General del Derecho. Santa Fé de Bogotá – Colômbia: Editorial
Temis.
CABRERA, C.A. [2000] Lecciones de lógica jurídica. Espanha: Editorial Mad, S.L.
GOYARD-FABRE, S. [1993] Kelsen e Kant. Saggi sulla dottrina pura del diritto. Napoli:
Edizioni Scientifiche Italiane.
GUERRIER, O. [2004] Fictions du Droit et Espace Littéraire. In Fictions du savoir à la
Renaissance, URL : http://www.fabula.org/colloques/document88.php
HARE, R. W. [1996]A Linguagem da Moral. Tradução de Eduardo Pereira e Ferreira. São
Paulo: Martins Fontes.
HUME, D. [2000] Tratado da Natureza Humana. São Paulo: Fundação Editora da UNESP
(FEU).
KAUFMANN, E. [1992] Critica della filosofia neokantiana del diritto. Napoli: Edizioni
Scientifiche Italiane.
KELSEN, H. [2005] Teoria Geral do Direito e do Estado. São Paulo: Martins Fontes, [1° ed.
1945].
______. [1962] Théorie Pure du Droit. Paris : Dalloz, [1° ed. 1960].
______. [1996] Théorie Générale des Normes. Paris : Léviathan, [1° ed. 1979].
LOEWENBERG, J. [1912] Reviewed work(s): Die Philosophie des Als Ob. System der
Theoretischen, Praktischen und Religiösen Fiktionen der Menschheit auf Grund Eines
Idealistischen Positivismus. Mit einem Anhang über Kant und Nietzsche by H. Vaihinger.
203
The Journal of Philosophy, Psychology and Scientific Methods, Vol. 9, No. 26 (Dec. 19,
1912). Published by: Journal of Philosophy, Inc. Stable URL:
http://www.jstor.org/stable/2013048
PFERSMANN, O. [2004] Les modes de la fiction : Droit er Littérature. In : LAVOCAT, F.
Usages et théories de la fiction, PUR.
ROSS, Alf. [2003] Direito e Justiça. Tradução Edson Bini. Bauru, SP: EDIPRO.
THOMASSON, A. [1999] Fiction and Methaphysics. Cambridge University Press,
Cambridge.
WICKER, G. [1997] Les fictions juridiques. Contribution à l’analyse de l’acte juridique.
Bibliothèque de Droit Privé Tome 253. L.G.D.J. Paris.
204
Normes juridiques et fictions littéraires:
confrontations entre Kelsen et Thomasson
Juliele Maria Sievers
Université Lille 3
UMR 8163 (Savoirs, Textes, Langage)
Dans ce texte on défend la thèse selon laquelle le domaine normatif juridique n’est pas
convenable à la présence des fictions. Dans une première vue, ce positionnement peut
sembler à la fois trop radical comme aussi banal. Il peut sembler radical parce que la présence
des fictions de type normatif, dans le domaine du droit, peut être attestée depuis les fictio
legis du Droit Romain, demeurant incontestable dans le quotidien de la pratique juridique.
Banal, parce que c’est considérablement claire le caractère problématique de ce type de
fiction, c'est-à-dire que tous sont d’accord sur les problèmes concernant la légitimation de son
usage. Les obstacles qui entourent cette légitimation ne semblent pourtant pas empêcher
l’apparent succès de l’usage de la fiction dans le domaine du droit.
Notre objet d’étude sera une des notions centrales de la théorie du droit proposée par le
juriste autrichien Hans Kelsen (1881-1973), celle de norme fondamentale, une fiction selon
cet auteur. On montrera que la norme fondamentale n’est pas une norme juridique comme
les autres normes présentes dans un ordre. En revanche, on essaiera de montrer qu’elle est
une fiction scientifique, un outil méthodologique à la disposition de la science du droit. Avec
cela, on montrera enfin que les normes ne sont jamais des fictions, ceci en raison aussi de la
clause kelsénienne qui pose l’efficacité d’une norme comme une des conditions pour sa
validité.
205
Dans la deuxième partie de notre texte, on investiguera quelques caractéristiques des fictions
littéraires selon la théorie artéfactuelle d’Amie Thomasson. Selon notre analyse de cette
théorie, on constate que les fictions littéraires auraient les mêmes caractéristiques des
normes juridiques, ce qui représenterait une forte attaque à notre approche. On argumentera
contre cette identification en analysant, d’un côté, les actes de création des personnages
fictifs littéraires et, de l’autre, les actes de création des normes juridiques. Notre investigation
montrera que la notion de fiction utilisée par le droit et celle utilisée par la littérature ne sont
pas de même nature. On ira même plus loin, en montrant que toute utilisation du terme
« fiction » dans le champ normatif juridique représente en fait un mauvais usage de ce terme.
1. La fiction dans le Droit
1.1 La norme fondamentale : norme ou fiction ?
L’objectif de ce texte c’est de montrer que le domaine normatif du droit n’est pas un champ
légitime pour la présence de la fiction. Pourtant, Kelsen nous présente comme notion centrale
de sa théorie juridique précisément une fiction, la norme fondamentale. Alors, comment
peut-on attester l’importance de cette « norme » et, au même temps, attaquer la présence
de la fiction dans le droit ? Commençons pour bien comprendre qu’est-ce que la norme
fondamentale, et pourquoi elle est une fiction.
La définition de norme fondamentale chez Kelsen est fréquemment donnée par rapport à sa
fonction spécifique dans le cadre d’un ordre juridique positif. De cette manière, Kelsen
explique que la norme fondamentale n’est pas dirigée vers une conduite déterminée, et
qu’elle n’impose pas une sanction, parce que son « destinataire » c’est l’ordre juridique lui-
même, dans son ensemble.
Selon Kelsen, la pluralité des normes juridiques appartenant à un seul ordre peut être
considérée comme étant organisée selon une pyramide. On part de la norme individuelle
« créée » dans le tribunal par le juge, qui sert de base à la pyramide. Mais cette norme
individuelle est en vérité aussi l’application d’une norme plus générale, de contenu plus
compréhensif, qui est déjà présente dans l’ordre, et qui serait, dans notre pyramide, au-dessus
de la norme qui vient d’être « créée ». Cette norme plus générale est à son tour basée sur une
206
norme encore plus générale, et de cette manière on « monte » dans la pyramide jusqu’à la
Constitution d’un Etat, par exemple. Il s’agit de l’autoproduction du droit positif, c'est-à-dire
qu’une norme, pour être une norme valide, doit obéir à deux procédures pour sa création : sa
validité doit être le produit d’un acte de volonté objectif (c'est-à-dire, cet acte doit être
effectué par une personne autorisé par le droit lui-même), et son contenu doit avoir un
correspondant dans une autre norme, plus générale, déjà présente dans l’ordre (c'est-à-dire
une norme déjà valide, une norme positive).
L’élaboration de la pyramide pour attester la légitimation de la validité de l’ordre constitue
une procédure de justification des démarches juridiques, c'est-à-dire un travail du juriste.
Alors, le juriste tombe sur un grave problème : la Constitution n’a aucune norme plus générale,
plus « haute » dans la pyramide, pour justifier sa validité. Le législateur a produit un acte de
volonté pour la création de la Constitution, mais il n’a pas été autorisé par une norme plus
« haute », et aussi, aucune norme plus « haute » n’a pas donnée le contenu normatif (aucune
norme plus « haute » n’a été appliquée). Ainsi, la Constitution a été produite sans appel à
aucun élément déjà existant dans l’ordre : elle a été le premier élément de cet ordre.
Pour résoudre ce problème, Kelsen présente la notion de norme fondamentale. Il s’agit d’une
fiction qui sert à légitimer l’existence de la Constitution et, indirectement, de tout l’ordre
juridique qui se suit. L’appel à la norme fondamentale dans le cadre d’un système juridique
positif est, selon Kelsen, indispensable parce que, sans elle, l’ensemble multiple de normes
ne pourrait pas être l’objet de la justification rationnelle donnée par la science juridique, c'est-
à-dire que la multiplicité des normes ne pourrait pas être vue comme un objet cognoscible.
De cette manière, la norme fondamentale est l’objet non pas d’un acte de volonté de la
personne autorisée, mais elle est le produit d’un acte de pensé de la personne intéressé à
comprendre ou à expliquer le droit. Or, le produit d’un acte de pensé n’est pas une norme.
Seuls les produits d’actes de volontés sont normes ou commandements, qui ne sont pas des
éléments du domaine de l’être, mais du devoir-être. Une conséquence importante c’est que
la norme fondamentale ne sera jamais une norme efficace, c'est-à-dire une norme respectée
ou appliquée dans la réalité du monde. La norme fondamentale est le fondement de la validité
d’un ordre normatif dans son ensemble. Elle est une fiction élaborée dans le cadre de la
science du droit.
C’est à ce point-là qu’on arrive à une question centrale: si la norme fondamentale ne peut pas
être observée, si elle « n’intéresse pas » à nous, mais seulement à la science du droit (qui
décrit les faits, mais ne pose pas de prescriptions), pourquoi appelle-t-on « norme » à la norme
207
fondamentale ? La réponse indique une raison purement terminologique : seules les normes
produisent des normes, et il faut une norme pour légitimer la création de la Constitution,
même si c’est une norme présupposée, une fiction. Si la « norme » fondamentale ne serait
pas appelée « norme », elle n’aurait pas le besoin d’être appelée aussi « fictive ». Mais elle ne
pourrait pas non plus atteindre à son but de justifier la création de la Constitution.
Dans la Théorie Pure du Droit200, la notion de norme fondamentale est vue par Kelsen comme
une présupposition. Mais dans sa dernière œuvre, la Théorie Générale des Normes201, Kelsen
dira que la norme fondamentale est une fiction. Or, si Kelsen dit que c’est le droit qui produit
le droit, et si le droit semble soudain être produit par une fiction, les choses deviennent
confuses. Mais la norme fondamentale ne produit pas le droit, elle justifie la création
normative. Et alors on peut déjà comprendre pourquoi on doit appeler « norme » à la norme
fondamentale, même si c’est clair le fait que la norme fondamentale ne partage pas les mêmes
caractéristiques des autres normes juridiques.
Pourtant, une critique toujours courante à respect de la norme fondamentale kelsénienne dit
que tout le système normatif positif ne pourrait pas être fondée sur une fiction, c'est-à-dire,
quelque chose qui n’est pas une norme positive. Mais on comprend ici que la fonction de la
norme fondamentale n’est pas de fonder le système dans le sens d’être la première norme
d’un ordre, mais que sa fonction est de justifier l’ensemble de cet ordre, dans un sens
méthodologique et scientifique. On soutient dans ce texte qu’aucune fiction ne peut avoir un
caractère normatif.
La norme fondamentale ne peut pas non plus être appelée une norme de deuxième degré, ou
un méta norme. Cela supposerait que la science du droit possède une fonction prescriptive,
c'est-à-dire qu’elle serait capable de produire le droit. Il est important de ne pas mélanger ces
deux domaines, ni les fonctions spécifiques de chacun d’eux: le droit produit les normes, le
droit c’est les normes juridiques ; la science du droit c’est la description de l’ensemble des
normes juridiques. En fait, la notion de norme fondamentale ne trouve aucune justification
en dehors du contexte scientifique. Elle ne produit rien, mais sert à mettre le point final dans
la quête pour la justification normative, ce qui permet de considérer un ensemble de normes
comme un seul système normatif. Or, la notion de norme fondamentale devient
problématique seulement à partir du moment où on pose la question « mais comment une
norme positive (par exemple, la Constitution d’un Etat) pourrait être basée dans la validité
200Kelsen, H. (1962). 201Kelsen, H. (1996).
208
d’une notion qui n’est pas normative, qui n’est pas elle-même valide ? ». La validité des
normes de la Constitution ne sont pas basées dans la validité (fictive) de la norme
fondamentale. La légitimité de la Constitution est expliquée par la notion de norme
fondamentale. La Constitution a été positivée par le législateur comme un ensemble de
normes valides, qui donnent le point de départ pour la production normative. Mais c’est la
justification de cet ensemble de normes qui est donnée par la norme fondamentale. Ainsi, le
domaine du Droit c’est le domaine de la production normative, tandis que le domaine de la
science du droit c’est le domaine de la justification normative. Kelsen (1962, p. 203) dit :
« Cette norme fondamentale ne peut donc pas être une norme voulue, elle ne peut
notamment pas être voulue par la science du droit – c'est-à-dire par le juriste qui pratique la
science du droit –, mais d’autre part, cette norme fondamentale (plus exactement, son
énoncé) est logiquement indispensable pour fonder la validité objective des normes juridiques
positives. »
Enfin, Kelsen (1962, p.205), dit aussi :
« En formulant cette théorie de la norme fondamentale, la théorie pure du droit n’inaugure
absolument pas une nouvelle méthode de la connaissance juridique. Elle ne fait qu’amener à
la pleine conscience ce que tous les juristes font, le plus souvent sans s’en rendre compte,
lorsqu’ils conçoivent que les données qui ont été caractérisées ci-dessus ne sont pas des faits
déterminés par les lois de la causalité, mais les considèrent conformément à leur sens
subjectif, comme des normes objectivement valables, comme un ordre juridique normatif,
sans rapporter la validité de cet ordre à une norme supérieure, métajuridique, c'est-à-dire
posée par une autorité supérieure à l’autorité juridique, – en bref : lorsqu’ils conçoivent le
droit exclusivement comme droit positif. »
Donc, si la norme fondamentale kelsénienne n’est clairement pas une fiction juridique, mais
une fiction scientifique, on s’intéressera ensuite à quelques exemples de ce qui est
normalement appelé « fiction juridique » dans la pratique du droit.
1.2 Les soi-disant fictions juridiques
Dans cette section, concentrons-nous sur ces exemples de normes qui sont normalement
appelés « fictions » du droit. Notre but sera de montrer que la présence de ce type de
209
formulation normative faisant appel à la fiction n’est pas légitime dans le contexte prescriptif
du droit. Pour le faire, on analysera d’abord l’approche qui fait Otto Pfersmann202 de cette
question. Pfersmann dit que la fiction est seulement une des multiples formulations possibles
de la norme, et pourrait très bien être évitée par le moyen de l’usage d’une autre technique
de formulation normative.
Pfersmann (2004, p.43)203 considère la norme suivante, analysée dans le contexte du
règlement d’un club : « Seules les femmes ont accès à cette piscine, le maître nageur est une
femme au sens de ce règlement ». La fiction réside ici dans le fait qu’un événement impossible
fait partie de la formulation normative, à savoir, « le maître nageur est une femme ».
L’impossibilité caractérise le fictive.
Pfersmann explique que la norme a été formulée avec la présence de la fiction selon un choix,
c'est-à-dire qu’il s’agit seulement d’une technique juridique spécifique entre autres. Dans les
mots de l’auteur :
« Dans l’exemple de la piscine, il n’est pas dit qu’un homme est une femme, mais simplement
que, au cas où il existe un maître-nageur du sexe masculin, cette personne doit être traitée
comme une personne du sexe féminin au sens du règlement de la piscine de N***, en d’autres
termes, il lui est permis de pénétrer dans la piscine. Le même résultat aurait été obtenu, si le
règlement avait dit :
N : Seules les femmes sont admises dans cette piscine, à l’exception du maître-nageur en
charge de la sécurité des personnes et des installations.
Or, il s’agit simplement d’une autre formulation de la même norme utilisant une autre
technique. » (Pfersmann, 2004, p.54)
De cette manière, même si Pfersmann ne semble pas vraiment critiquer ou attaquer la notion
de fiction dans le Droit, il atteste tout au moins que sa présence n’est que simplement
optionnelle, c'est-à-dire, qu’elle peut toujours être évitée.
202 Pfersmann O. (2004). 203En fait, Pfersmann défend, dans son texte, la thèse selon laquelle le Droit serait, par nature, entièrement
fictif. C’est clair qu’on est contre cette approche. Pourtant, cette partie de son argumentation, à savoir
concernant le problème de la formulation normative avec la présence des fictions, est entièrement
compatible avec notre thèse contre la fiction dans le domaine normatif.
210
1.2.1 L’impossibilité d’efficacité d’une « fiction juridique »
On restera sur l’analyse du même exemple donné par Pfersmann, mais pour cette fois
présenter une forte attaque à la présence de la fiction dans le Droit. On défend que la fiction
« simplifie » la formulation de la norme, mais qu’elle n’est plus considérée dans le moment
de l’application de la norme. Parce qu’au moment de l’application ou de l’observation de la
norme « fictive », le contenu de la norme n’arrive jamais à être observé, parce qu’il est
impossible que le maître-nageur « devienne » une femme. Quand on applique une norme qui
présente des fictions dans sa formulation, c’est exactement sa formulation alternative, sans
la fiction, qui est en effet considérée.
C’est une thèse de la théorie kelsénienne que l’efficacité d’une norme a une place centrale
dans la détermination de la validité normative. A savoir, Kelsen dit qu’une norme doit être
dans un minimum efficace (c'est-à-dire observée ou encore appliquée) pour continuer à être
valide (c'est-à-dire pour continuer présente dans un ordre juridique positif). Cela ne veut pas
dire que toutes les normes valides sont efficaces, justement parce qu’une norme seulement
devient efficace après être valide, après exister comme norme, après être présente dans un
ordre. Le point c’est que, si la norme n’est pas minimalement efficace, elle ne sera plus valide.
Or, c’est le moment de se demander : quand est-ce qu’une fiction sera respectée dans la
réalité du monde par les destinataires de la norme ? Jamais. Jamais un maître-nageur ne sera
une femme. La seule chose qui peut vraiment arriver c’est que le maître-nageur soit traité, lui
seul, de la même manière que sont traitées les femmes, et il n’y a rien de fictif ou d’impossible
dans cela.
C’est important de souligner qu’on n’est pas en train de dire que nous « réalisons » la norme
au moment de son observation. Cela représenterait un attentat à la thèse kelsenienne de
l’abîme méthodologique entre les domaines de l’être et du devoir-être. La norme ne devienne
pas réelle quand elle est appliquée, mais son contenu, lui, est effectivement accompli par
notre comportement. Et c’est exactement dans la formulation normative, dans son contenu,
que la fiction est présente.
Ainsi, on termine cette section avec deux constatations importantes : la fiction présente dans
le droit c’est celle qui impose une condition absurde, une impossibilité, comme un fils adoptif
qui est vu comme fils légitime ou une entreprise qui est considérée comme une personne. Or,
211
c’est très simple de constater qu’une fiction, dans ce sens d’impossibilité, ne peut pas être
observée par le moyen d’un comportement spécifique des destinataires de la norme. Et alors,
selon Kelsen, la norme qui n’est pas minimalement efficace dans la réalité du monde perd son
statut de norme positive. Par cette voie, donc, on défend que toute présence de normes qui
ont des fictions dans son formulation représente une « erreur catégorique », un mauvais
usage d’un terme non légitime aux démarches prescriptives du droit. En plus, cette erreur
peut être évitée simplement par une différente formulation de la norme, comme le soutient
Pfersmann. La fiction comme celle du « maître-nageur qui est une femme », a un excellent
pouvoir d’expliquer la norme, de simplifier sa formulation. Ce pouvoir explicatif peut sans
problème être profité par le droit, mais il sera placé dans la science juridique, le domaine
propre aux explications et justifications des démarches juridiques.
2. La fiction dans la littérature
Dans cette section, on analysera que la fiction littéraire n’est pas forcément identifiée avec
des impossibilités, comme on a vu jusqu’ici avec nos exemples. Dans le domaine littéraire, il y
a des fictions qui sont bien possibles dans la réalité, et c’est peut-être en raison de ce fait
qu’on trouve certaines histoires de Poe si effrayantes, par exemple. Contrairement à un
« maître-nageur qui est une femme », on pourrait parfaitement rencontrer un chat noir sans
un œil dans la rue… Dans cette deuxième partie de notre texte, on défendra que le critère
d’identification de la fiction littéraire ne dépende pas de sa modalité, mais de l’intentionnalité
de son auteur. Cela pose un obstacle très solide à l’identification tracée par Thomasson204, qui
fait partager les mêmes caractéristiques des fictions littéraires aux normes. Or, si la fiction
normative consiste déjà en une erreur, encore plus grave c’est son identification avec les
fictions littéraires, qui d’ailleurs n’ont même pas de contraintes modales.
Thomasson commence à nous donner des pistes pour sa définition de fiction205 ou, plus
précisément, d’un objet fictionnel, quand elle dit :
204 Thomasson A. (1999). 205 En fait, la définition de fiction littéraire donnée par Thomasson dans « Fiction and Metaphysics » est élaborée à partir d’un complexe système construit en termes de dépendances ontologiques. Mais, dans notre texte, il nous est déjà suffisante sa conception de création de la fiction faite par des actes de créativité, pour montrer déjà la différence entre l’acte de création en littérature et celui en droit.
212
“Just as marriages, contracts, and promises may be created through the performance of
linguistic acts that represent them as existing, a fictional character is created by being
represented in a work of literature. If there is no preexistent object to whom Austen was
referring in writing the words above, writing those words brings into existence the object
therein described: The fictional character Emma Woodhouse.” (Thomasson, 1999, p.13).
Le problème se montre dans le fait que ces mêmes aspects peuvent être observés dans le
domaine juridique. En effet, une norme trouve son existence par la performance d’un acte
linguistique qui représente son existence comme un élément du droit. Mais le point c’est qu’il
s’agit d’un type très particulier d’acte linguistique : Tandis que l’acte qui « fait exister » un
objet fictionnel est spécifique au domaine littéraire, l’acte que « fait exister » une norme est
aussi et déjà un élément propre du droit.
Selon notre approche qui vise différencier complètement les normes des fictions de tout
espèce et, en particulier ici, des fictions littéraires, il faut faire une différenciation entre les
actes de création de chaque domaine. L’acte de création propre de la littérature ce sont des
actes d’imagination. L’acte de création d’une norme juridique est un acte de volonté, et plus
précisément, l’acte de volonté objectif, réalisé par l’autorité en question. En plus, au contraire
de ce qui se passe en littérature, l’autorité de l’auteur de l’acte de volonté est déjà un élément
propre du droit : l’autorité de l’auteur, du « créateur », est un aspect que seul le droit peut
concéder, qui n’est pas constitutif de la personne en question, mais du domaine où son
pouvoir, son autorité, sont pratiquées.
Donc, une analyse des différents actes de création d’objets littéraires et de normes juridiques
montre un aspect central pour montrer que les normes ne sont jamais des fictions : dans la
littérature, l’acte de créativité dépend profondément de l’intention de l’auteur de la fiction.
L’auteur créé le personnage comme étant une fiction. Par contre, dans le droit, l’acte de
volonté objectif ne peut créer que des normes, parce qu’on est dans un champ limité par des
aspects purement juridiques. C'est-à-dire que l’auteur de la norme n’est autorisé par le droit
que pour produire/appliquer des normes juridiques. Son acte de volonté n’a pas de portée
au-delà de cette action. Dans le droit, le juge ou législateur sont limités par la norme elle-
même qui les a attribuées ses pouvoirs. Ainsi, on comprend que la clé pour différencier les
fictions littéraires des normes juridiques c’est l’analyse des actes de création de ces deux
objets.
213
Dans le cas de la norme fondamentale, c’est maintenant très claire la différence : La norme
fondamentale n’est pas créée, elle est présupposée. Et il n’est même pas l’autorité du droit
qui la présuppose, mais le juriste, le scientifique. C’est pour cela qu’elle demeure une fiction
non pas du droit, mais pour le droit.
Conclusion
L’objectif de ce texte c’été de défendre la thèse selon laquelle le domaine normatif juridique
n’est pas convenable à la présence des fictions. Pour le faire, on a montré qu’un aspect central
qui empêche cette présence c’est la nécessité de l’application d’une norme pour qu’elle
continue à être une norme juridique valide ou valable. Or, comme on a vu avec les exemples
qui ont été analysés, il est impossible d’observer ou appliquer une fiction juridique.
On a montré qu’une fiction juridique comme, par exemple, celle que dit que « L’enfant conçu
est tenu pour né chaque fois qu’il y va de son intérêt » est impossible d’être observé dans la
réalité, parce que l’enfant conçu ne serait pas un enfant né. On a défendu ici que cette
formulation de la norme doit être vue comme une explication de la même norme, donné par
la science du droit pour expliquer pourquoi la même norme destiné à l’enfant né est
maintenant en train d’être aussi destinée à l’enfant conçu. C’est une erreur de placer cette
norme dans le champ juridique, parce qu’elle ne pourra jamais être observée tel qu’elle est
formulée, c'est-à-dire que son contenu ne sera jamais abouti par ses destinataires.
Dans la deuxième partie, on a présenté des arguments contre l’identification de la norme avec
les fictions littéraires. La théorie artéfactuelle d’Amie Thomasson s’est présentée comme un
contexte où cette approche semble avoir lieu. Notre point de départ pour le critiquer c’été
l’analyse des actes de création d’une fiction littéraire et l’acte de création d’une norme
juridique. On a vu que la création d’une norme ne passe pas par les mêmes démarches que
l’invention d’un personnage littéraire. Dans le contexte du droit, l’acte de volonté doit être
effectué par une personne autorisée par le droit lui-même et, en plus, cet acte doit être
soutenu par le fait de l’existence d’une norme de contenu plus général, déjà présente dans
l’ordre normatif (la création de la nouvelle norme est toujours l’application de la norme plus
générale).
214
Finalement, on a montré que la norme fondamentale, entant que fiction, n’est pas de même
nature que les normes juridiques « ordinaires ». Elle s’agit d’une fiction scientifique, utilisée
avec le but précis de légitimer la création normative. La norme fondamentale n’a pas le besoin
d’être observée ou appliquée, et c’est pour cela, pour être une norme impossible, qu’elle ne
peut pas, comme aussi n’importe quelle autre fiction, être placée dans le domaine normatif.
Ainsi, si la pratique du droit est marquée par la présence des ainsi appelées « normes fictives »,
c’est à cause du apparent succès de cette pratique dans la simplification des démarches
juridiques. Néanmoins, au moment de l’application de ces normes, ce n’est pas la formulation
fictive qui est considérée, mais son formulation alternative, sans appel à la fiction. La fiction
peut avoir un pouvoir explicatif dans le sens de nous faire mieux comprendre une loi, mais
nous ne pouvons pas réaliser la fiction par le moyen de nos actions et comportements. Ce
type de pouvoir, d’obliger ou interdire pleinement une conduite spécifique, appartient aux
normes de la morale ou du droit et, pour le dernier, il faut absolument que ses normes soient
possibles d’être respectées ou même violées par ses destinataires.
Références Bibliographiques :
Kelsen, H. [1962] : "Théorie Pure du Droit". Paris : Dalloz, [1° éd. 1960].
Kelsen, H. [1996] : "Théorie Générale des Normes". Paris : Léviathan, [1° éd. 1979].
Pfersmann O. [2004] : "Les modes de la fiction : Droit er Littérature". In : Lavocat, F. Usages
et théories de la fiction, PUR.
Thomasson A. [1999]: "Fiction and Metaphysics". Cambridge University Press, Cambridge.
215
Bibliography
Aquinas, T. (2014). Summa Theologiae I-II qq. 90-106. The Summa Theologica: Complete
Edition. Catholic Way Publishing. New York. [1265–1274].
Aristotle. Rethoric. Book 1: Chapter 13. [1373b] Austin, J. (1995). The Province of Jurisprudence Determined. W. Rumble (Eds.). Cambridge
University Press. Cambridge [1832].
Bindreiter, U. (2002). Why Grundnorm? A treatise on the Implications of Kelsen’s Doctrine. In:
Kluwer Law International. Law and Philosophy Library. Neatherlands.
Bobbio, N. (2008). Direito e Poder.Translation of Nilson Moulin. Ed. UNESP. São Paulo.
Bobbio, N. (1997). Teoría General del Derecho. Editorial Temis. Santa Fé de Bogotá, Colômbia.
Bulygin, E. (1990). An Antinomy in Kelsen’s Pure Theory of Law. In: Ratio Juris, 3. Pp. 29-45.
Cabrera, C. A. (1999). Imperativos y lógica en Jørgen Jørgensen. In: Isegoría, 20. Pp. 207–215.
Calhoun, G. C. (1944) Xenophon, Memorabilia, I. ii. 40-6, trans. George C. Calhoun. In: Calhoun, Introduction to Greek Legal Science (Oxford: Clarendon Press, 1944, repr. Scientia: Aalen, [1977]). Celano, B. (2000). Kelsen’s Concept of the Authority of Law. In: Law and Philosophy, Vol. 19,
No. 2, Mar, p. 173-199.
Champeil-Desplats, V. (2002). Alf Ross: droit et logique. In : Droit et société, (1). Pp. 29–42.
Clerbout, N. (2013). Étude de quelques sémantiques dialogiques. Concepts fondamentaux et
éléments de métathéorie. PhD Thesis, Université de Lille.
Curry, O. (2006). Who’s Afraid of the Naturalistic Fallacy? In: Evolutionary Psychology, 4. Pp.
234-247.
216
Beyleveld, D. & Brownsword, R. (1998). Methodological Syncretism in Kelsen’s Pure Theory of
Law. In: Norms and Normativity. Stanley and Bonnie Paulson (Eds.).Clarendon Press Oxford.
Gabbay, D. & Woods, J. (2010). Logic and the Law: Crossing the Lines of Discipline. In:
Approaches to Legal Rationality. Gabbay, Canivez, Rahman, Thiercelin (Eds.). Logic,
Epistemolog, and the Unity of Science. Volume 20. Springer.
Hansen, J., Pigozzi, G. and van der Torre, L. (2007). Ten philosophical problems in deontic logic.
In: Proceedings of Normative Multi-agent Systems, NORMAS 2007.
Hare, R. M. (1972). The language of morals. Oxford University Press. Oxford.
Hart, H. (1998). Kelsen visited. In: Norms and Normativity. Paulson, S. & Paulson, B. (Eds.).
Clarendon Press Oxford, p. 70.
Hart, H. (1983). Kelsen Visited. In: Essays in Jurisprudence and Philosophy. Oxford: Clarendon
Press, 286-308.
Heidemann, C. (1997). Die Norm als Tatsache: Zur Normentheorie Hans Kelsens. Baden-
Baden. Nomos.
Holmes, O.W. (1881). The Common Law. Available at:
http://www.gutenberg.org/files/2449/2449-h/2449-h.html
Hume, D. (2007). A Treatise of Human Nature. Reprinted from the Original Edition in three
volumes by L.A. Selby-Bigge, M.A. Oxford: Clarendon Press. [1739].
Kammerhofer, J. (2005). Unearthing Structural Uncertainty through Neo-Kelsenian
Consistency: Conflicts of Norms in International Law. In: Papers of the European Society of
International Law. December 2005.
Kant, I. (2005). Groundwork of the Metaphysics of Morals. Routledge, New York. [1785].
Kant, I. (1998). Critique of Pure Reason. Cambridge University Press. Cambridge. [1781].
Keiff, L. (2009). Dialogical logic. http://plato.stanford.edu/entries/
Kelsen, H. (1923). Haupteprobleme der Staatsrechtslehre. 2nd printing (Tübingen: J. C. B.
Mohr).
Kelsen (1960). Pure Theory of Law. Translated from the German by Max Knight. University of
California Press, 1989.
Kelsen, H. (1979). General Theory of Norms. Clarendon Press, Oxford, 2001.
Kelsen, H. (1998). The Pure Theory of Law, ‘Labandism’ and Neo-Kantism. A Letter to Renato
Treves. In: Norms and Normativity. Paulson, S. & Paulson, B. (Eds.). Clarendon Press Oxford,
p.169-175.
217
Kelsen, H. (2008). What is Justice? Justice, Law and Politics in the Mirror of Science. Collected
Essays. University of California Press, Berkeley, Los Angeles [1957]. Repr. The Lawbook
Exchange, Clark, New Jersey.
Loewenberg, J. (1912). Review. In: The Journal of Philosophy, Psychology and Scientific
Methods, Vol. 9, No. 26 (Dec. 19, 1912).
Lorenzen, P. and Lorenz, K. (1978). Dialogische Logik. Wissenschaftliche Buchgesellschaft,
Darm- Stadt.
Magnier, S. (2012). PAC vs. DEMAL, A Dialogical Reconstruction of Public Announcement Logic
with Common Knowledge. In: Logic of Knowledge. Theory and Applications. C. Barès, S.
Magnier, and F. Salguero (Eds.). College Publications, London. Pp. 159–179.
Magnier, S. (2013). Approche dialogique de la dynamique épistémique et de la condition
juridique. College Publications, London.
Magnier, S. & de Lima, T. (2014). A Soundness & Completeness Proof on Dialogs and Dynamic
Epistemic Logic. P. Allo, F. Poggiolesi, and S. Smets (Eds.). In: Dynamics in Logic. Logique et
Analyse.
Mangier, S. & Rahman, S. (2012). Leibniz’s Notion of Conditional Right and the Dynamics of Public Announcement. In: Limits of knowledge society, D.G. Sambotin (Ed.), pp. 87-103.
Natural Law for Today’s Lawyer. In: Stanford Law Review, Vol. 9, No. 3, May, 1957, p. 479. Ott, W. (1987). Bericht von einem Besuch bei Prof. H.L.A. Hart in Oxford. Duncker & Humblot.
Berlin.
Paulson, S. (1992). ‘The Neo-Kantian Dimension of Kelsen’s Pure Theory of Law’. In: Oxford
Journal of Legal Studies, 12. Oxford University Press.
Paulson, S. (1998). Four Phases in Hans Kelsen’s Legal Theory? Reflections on a Periodization.
In: Oxford Journal of Legal Studies, 18. Oxford University Press. Pp. 153-166.
Paulson, S. (1999). Arriving at a Defensible Periodization of Hans Kelsen’s Legal Theory. In:
Oxford Journal of Legal Studies, 19. Oxford University Press. Pp. 351-364.
Pfersmann, O. (2004). Les modes de la fiction. Droit et littérature. In : Usages et théories de la
fiction, Le débat contemporain à l'épreuve des textes anciens. Françoise Lavocat (Dir.). Rennes,
Presses Universitaires de Rennes. Pp. 39-61.
Poincaré, H. (1982). The Foundations of Science: Science and Hypothesis, The Value of
Science, Science and Method. University Press of America. [1913].
Rahman, S. (2006). Non-normal dialogics for a wonderful world and more. In: The Age of
Alternative Logics. Pp. 311–334.
Rahman, S. & Rückert, H. (1999). Dialogische Modallogik (für T, B, S4, und S5). In: Logique et
218
Analyse, 167(168). Pp. 243–282.
Rahman S. & Tulenheimo, T. (2009). From games to dialogues and back. In: Games: Unifying
Logic, Language, and Philosophy. Pp. 153–208.
Reynolds, B. (1993). Natural Law versus Positivism: The Fundamental Conflict. In: Oxford Journal of Legal Studies. Vol. 13, No. 4, Winter.
Ross, A. (1944). Imperatives and logic. In: Philosophy of Science, 11(1). Pp. 30–46.
Ross, A. (1961). Validity and the Conflict between Legal Positivism and Natural Law. In: Revista
Juridica de Buenos Aires, n°4. Pp. 46-93.
Sartor, G. (2005). Legal Reasoning: A Cognitive Approach to the Law. In: Treatise on Legal
Philosophy and General Jurisprudence, Vol. 5. Springer, Berlin.
Shapiro, S. J. (2009). What is the Rule of Recognition (And does it Exist)? In: Public Law & Legal
Theory Research Paper Series. Research Paper No. 181. YALE Law School.
Sievers, J. (2009). “Kelsen sobre o Lugar da Lógica no Âmbito Normativo. (Master’s
Dissertation).
Available at: http://w3.ufsm.br/ppgf/menuesp2/404d5d77d7e8468e2c0e3fafa65fdb4f.pdf
Sievers, J. (2011). Peut-on respecter une norme impossible? In : Normes et Fiction. Cahiers de
Logique et Epistemologie. College Publications. London, pp. 123-140.
Sievers, J. & Rahman, S. (2011). Normes et Fiction. Cahiers de Logique et Epistemologie.
College Publications. London.
Sievers, J. & Magnier, S. (2015). Reasoning with Form and Content. In: Past and Present
Interactions in Legal Reasoning and Logic. Armgardt, M., Canivez, P. & Chassagnard-Pinet, S.
(Eds.). Logic, Argumentation & Reasoning, Vol. 7. Springer. (In print).
Silving, H. (1955). The Twilight Zone of Positive and Natural Law. In: California Law Review, Vol. 43, No. 3, Jul., p. 489. Spaak, T. (2005). “Kelsen and Hart on the Normativity of Law”. In: Perspectives on
Jurisprudence: Essays in Honour of Jes Bjarup. Peter Wahlgren (Ed.). Pp. 397-414.
Student Author (2002). Lessons from Abroad: Mathematical, Poetic, and Literary Fictions in
the Law. In: Harvard Law Review.Vol. 115, No. 8 (Jun., 2002). Pp. 2228-2249.
Styron, W. (1987). Sophie’s Choice. Corgi, London [1979].
Vaihinger, H. (1911). The Philosophy of “As-If”: a system of the theoretical, practical and
religious fictions of mankind. London, Routledge & K. Paul, 1965.
van Ditmarsch, H., van der Hoek, W. and Kooi, B. (2008). Dynamic Epistemic Logic. In: Synthese
Library: Studies in Epistemology, Logic, Methodology, and Philosophy of Science. 337. Springer,
Dordrecht.
219
van Roermund, B. (2000). “Authority and Authorization”. In: Law and Philosophy, Springer,
Vol. 19, No. 2. Pp. 201-222.
von Wright, G. H. (1951) Deontic logic. In: Mind, 60(237). Pp. 1–15.
von Wright, G. H. (1957). Logical studies. Routledge and K. Paul, London.
Winkel, L. (2004). Peace Treaties and International Law in European History. Randall Lesaffer
(Ed.). Cambridge University Press.
Wolfe, C. (2003). Understanding Natural Law. In: The Good Society. Penn State University Press, Vol. 12, No. 3. Wood, N. (1988). Cicero's Social and Political Thought. University of California Press. Berkeley
and Los Angeles, California.
Xenophon, Memorabilia, I. ii. 40-6, trans. George C. Calhoun. In: Calhoun, Introduction to
Greek Legal Science (Oxford: Clarendon Press, 1944, repr. Scientia: Aalen, 1977), Pp. 78-80.
220
Abstract
The present work offers an original approach on the legal notions developed by Hans Kelsen
in his attempts towards a “pure” theory of Law, based on a philosophical analysis of the main
legal concepts that have a strong philosophical feature, namely those notions which are
somehow “shared” between the two fields in their name, but not always in their meaning.
While the most striking notion to be approached via a philosophical perspective would
probably be that of legal validity (since validity is a central term also in Logic), we aim, in the
same way, to approach the notions of legal fictions, the notion of science in Law, normative
conflicts or “contradictions” as they are commonly – and wrongly – named, and the rule of
inference as it is applied in the context of normative creation, giving place to the wrong notion
of practical reasoning. The notion of practical reasoning is very rich in this context of
comparison, and will be a special one, as it serves for us to analyze traditional problems of
legal theory, such as Jørgensen’s dilemma, as well as it offers us the opportunity of providing
our own alternative of a logical treatment of the process of legal justification of the creation
of a norm. We aim to analyze the notion of legal and logical conditions as well, which
represent a changing in Kelsen’s perspective on the utility and legitimacy of the application of
logic to the legal domain.
Such a comparative study, even if it appears to be fundamental for clarifying those notions in
their respective fields, is a task never before developed in this systematic manner.
The objective of such a study is to provide a clear overview of the boundaries between the
fields of philosophy (especially logic) and the legal norms. A clear understanding of the
relations between those “homonym” notions may explain why they are most of the time
misused when philosophers talk about law, as well as when lawyers try to justify the concepts
composing the legal theory.
The context of this study is the legal positivism as it is explained by the legal-philosopher Hans
Kelsen. This choice is justified by the fact that Kelsen’s legal theory appears to be the most
221
suitable frame for an analytical, logic-oriented investigation. The work emphasized will be the
General Theory of Norms (1979), mainly because of the fact that this book represents how
intensively Kelsen dedicated himself to the legal problems mostly related to philosophy or
logic, namely the question of the application of logic to norms and the clarification of
problematic notions such as the basic norm as a fiction or, still, the notion of modally
indifferent substrate.
222
Résumé
Le travail ici présenté offre une approche originelle des notions juridiques développées par
Hans Kelsen dans ses efforts vers l’élaboration d’une théorie « pure » du droit. Cette étude
est basée sur une analyse philosophique des principaux concepts juridiques qui ont, à leur
tour, des caractéristiques philosophiques pertinentes, principalement les notions qui sont
apparemment « partagées » entre les domaines juridique et philosophique, sans pour autant
garder les mêmes significations sémantiques. La notion la plus pertinente à avoir ces
caractéristiques est peut-être celle de « validité », étant donné qu’il s’agit d’une notion
présente tant dans le domaine de la logique quant dans le domaine juridique. Dans ce sens,
nous prétendons approcher aussi les notions de fiction normative, de science du droit, de
conflits entre normes (qui sont aussi erronément appelés « contradictions entre normes »),
et la règle d’inférence comme elle est appliquée dans le contexte de la création normative,
donnant lieu à des notions trompeuses comme celle de raisonnement pratique. La notion de
raisonnement pratique est d’ailleurs très riche concernant ce contexte de comparaison, et elle
sera particulièrement traitée en fonction des autres notions problématiques qu’elle génère
dans le champ normatif, telles que le Dilemme de Jørgensen. Cette notion, étant employée
dans différentes cas problématiques du droit, nous offre une opportunité de présenter une
approche alternative pour le traitement logique du processus de justification dans la création
d’une norme. Nous avons pour objectif aussi d’analyser les notions de condition logique et
juridique, qui représentent un changement dans la perspective de Kelsen concernant l’utilité
et la légitimité de l’application de la logique dans le champ normatif du droit.
Une telle étude comparative, même si elle semble être fondamentale pour éclaircir les notions
en question dans ces domaines respectifs, reste une tâche jamais réalisée dans une telle
manière systématique.
L’objective de cette étude est de fournir un panorama clair concernant les limites entre les
champs de la philosophie (spécialement la logique) et les normes juridiques. La
223
compréhension des relations entre les notions « homonymes » doit aussi expliquer pourquoi
elles sont utilisées de forme erronée aussi fréquemment quand les philosophes discutent du
droit, mais aussi quand les juristes cherchent des justifications pour les concepts présents
dans leurs théories.
Ainsi, le contexte de cette étude est le positivisme juridique tel qu’il est présenté par le
philosophe du droit Hans Kelsen. Notre choix est justifié par le fait que la théorie de Kelsen
s’avère la plus conforme à notre approche philosophique d’orientation logique et analytique.
L’œuvre qui servira de base à notre investigation est la Théorie Générale des Normes (1979),
surtout dû au fait de représenter l’intensité des efforts de son auteur pour traiter les
problèmes juridiques liés à des aspects philosophiques ou logiques. Cela est particulièrement
remarquable dans le cas de la possibilité d’application de la logiques aux normes, mais aussi
concernant les notions problématiques comme celle de norme fondamentale comme fiction
ou encore la notion de substrat indifférent au mode.
224
Acknowledgements
This thesis represents the work performed at the Laboratoire Savoirs, Textes, Langage, at the
University of Lille 3, but also, in its final period, in the “Lehrstuhl für Bürgerliches Recht, Antike
Rechtsgeschichte, Römisches Recht und Neuere Privatrechtsgeschichte” in the University of
Konstanz. This exchange was made possible thanks to the French-German Project ANR-DFG
“JuriLog”, which I was glad to be a member during the three years of its existence, from 2012
to 2015. The researches I developed and conferences I made in the frame of this project, as
well as the experiences with the organization of its meetings and international conferences,
not to mention the exchanges with the professors and colleagues involved in the project, were
all fundamental for attaining the results I am able to present as a researcher today.
I would like to thank the Laboratory STL of Lille 3, it’s previous as well as its actual director,
professors Christian Berner and Philippe Sabot, who always supported my researches, either
by providing me a location allowing me to work in a fruitful environment together with my
colleagues at the Laboratory, either by helping with the funding of my travels concerning
scientific conferences where I took part, as well as for the publication of the book I edited
together with professor Shahid Rahman in 2011.
I would like to thank the Ecole Doctorale of Lille for the same kind of contribution to the
travels, as well as the collaboration during the preparation of the defense of the thesis and
the overall support during the time of the doctorate.
Similarly, I present my gratitude to the Maison Europeenne des Sciences de l’Homme et de la
Société (MESHS), which funded and gave me the opportunity to organize an international
seminar in the very first year of my doctoral studies. I’m also grateful for the constant support
of their personnel, especially Mme. Pascale Desquiens, Mme. Marie-Aude Depuiset and Mme.
Myriam Caudrelier.
225
I would like to thank my colleagues of the Law department of the University of Konstanz,
especially Karlheinz Huelser, Bettine Jankowski and Giuliano Bacigalupo. Their company and
collaboration, together with their friendship, were very important during the time I was a
member of the chair at that University.
My gratitude also to Prof. Matthias Armgardt, who welcomed and trusted me to make part of
his team in Konstanz. During the one year I worked near professor Armgardt I was enriched
with fruitful discussions, comments and advises wich now find an echo in the work I’m
presenting. Prof. Armgardt’s shared passion for Philosophy and Law certainly inspired me for
the interdisciplinary work I develop here.
I also thank my colleagues from the University of Lille, especially Nicolas Clerbout, Matthieu
Fontaine, Cristina Barés Gómez, Bernadette Dango and Virginie Fiutek. Their friendship and
support in working as well as in personal matters will never by forget. I would like to give my
especial thanks to my colleague Sébastien Magnier, with whom I collaborated to a great
extend, who worked with me in questions involving Law and Logic in the Project JuriLog and
who opened my eyes to the possibility of a correct using of logic for the legal field. I’m very
grateful for his patience on explaining me the logical features that I finally allowed in my vision
of the legal field. I’m also gratefull for his friendship that I sincerely hope to cultivate for the
rest of my life.
My most especial thanks are directed to the supervisor of this thesis, Prof. Shahid Rahman. I
have no words to express my gratitude for the support, the trusting, the encouragments and
the inspiration that he represents to me. It was an honour to be a doctorate student of Prof.
Rahman, which really represents, in my view, the scientific and philosophical profile and
attitude that I belive to be the ones to be followed in the academic context.
Last but not least, I would like to thank my friends and family that, even far from me, were
always present with their encouragements and support, but also with their conforting during
the difficult times. My thanks to the Arz family, which kindly “adopted” me and made France
look more like home. My special thanks also to my parents, Ademir and Iseli Sievers, and to
my dear brother Joel. My brother was, since I was born, my main encourager, the one who
always believed in me. It is to him that I dedicate this work.
I also thank the members of the Jury for their participation in this important moment of my
academic life, which I hope will be “long and prosper”. I therefore thank Prof. Patrice Canivez,
226
Prof. Sandrine Chassagnard-Pinet, Prof. Jan-Reinard Sieckmann, and again Prof. Matthias
Armgardt and Prof. Shahid Rahman.